Patent Application
20210407318
The disclosed embodiments relate generally to digital assistants, and more specifically to digital assistants that intelligently provide training opportunities and assistance to users.
Just like human personal assistants, digital assistants or virtual assistants can perform requested tasks and provide requested advice, information, or services. An assistant's ability to fulfill a user's request is dependent on the assistant's correct comprehension of the request or instruction. Recent advances in natural language processing have enabled users to interact with digital assistants using natural language, in spoken or textual forms, rather than employing a conventional user interface (e.g., menus or programmed commands). Such digital assistants can interpret the user's input to infer the user's intent; translate the inferred intent into actionable tasks and parameters; execute operations or deploy services to perform the tasks; and produce outputs that are intelligible to the user. Ideally, the outputs produced by a digital assistant should fulfill the user's intent expressed during the natural language interaction between the user and the digital assistant. A digital assistant can perform searches in a selection domain (e.g., a restaurant domain, etc.) and present qualifying selection items (e.g., restaurants) in response to a search request received from a user.
The ability of a digital assistant system to produce satisfactory responses to user requests depends on the natural language processing, knowledge base, and artificial intelligence implemented by the system. Conventional digital assistants respond to user commands of a single language, and provide responses in the same language. These digital assistants are not adequate when a user visits a region where a different language is used. In addition, although some conventional digital assistants respond to training or customization by a user, conventional digital assistants have not been useful in providing training to a user by leveraging its natural language and intent processing capabilities.
The embodiments disclosed herein provide methods, systems, a computer readable storage medium and user interfaces for a digital assistant to intelligently and proactively provide training opportunities and assistance to a user by leveraging its natural language processing and intent processing capabilities, particularly in foreign language training and assistance, and/or in introducing locale-specific accents/slangs to the user. An intelligent digital assistant with multi-lingual capabilities can be more effective in foreign language training because it does not simply perform a direct literal translation of the user's input; instead, the training samples and foreign language assistance are provided based on user intent inferred based on the user's input. Intent inference may be contextual, and can utilize relevant information about the user's current needs from many different sources. Many limitations of direct translation (e.g., mistakes due to awkward and incorrect sentence structure, grammar, and usage in the input) may be avoided. In addition, digital assistants are aware of the current context associated with the user, and may provide more appropriate foreign language training exercises to the user for the current context and provide motivation and real life practice opportunities for the user's learning.
Accordingly, some embodiments provide a method for operating a digital assistant, the method including, at a device including one or more processors and memory storing one or more programs: receiving, from a user, a first speech input spoken in a first language; inferring a user intent based on at least the first speech input in the first language; based on the inferred user intent, generating one or more alternative expressions of the first speech input in the first language; and providing feedback to the user introducing the alternative expressions as a more preferred input to express the inferred user intent than the first speech input provided by the user.
In some embodiments, providing the feedback further includes: providing the feedback in a second language different from the first language, wherein the second language is a primary language associated with the user, and the first language is a secondary language associated with the user.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that corrects a pronunciation of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that corrects a grammatical usage of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that replaces at least one word or phrase in the first speech input with another word or phrase.
In some embodiments, the method further includes: providing at least a command mode and a foreign language training mode, wherein the digital assistant executes a task flow to fulfill the inferred user intent in the command mode, and wherein the digital assistant generates the one or more alternative expressions and provides the feedback to the user in the foreign Language training mode.
In some embodiments, the method further includes: concurrently providing both the command mode and the foreign language training mode, wherein the digital assistant executes the task flow to fulfill the inferred user intent, in addition to generating the one or more alternative expressions and providing the feedback to the user.
In some embodiments, the method further includes: receiving user selection of the foreign language training mode; and enabling the foreign language training mode in response to the user selection of the foreign language training mode.
In some embodiments, the method further includes: automatically, without user intervention, enabling the foreign language training mode based on a current location of the user, wherein a primary language associated with the current location of the user is the first language.
In some embodiments, inferring the user intent based on the first speech input in the first language further includes: identifying a customized speech-to-text model of the first language for the user, wherein the customized speech-to-text model has been established based on training samples provided by native speakers of a second language (of which language the user is also a native speaker); processing the first speech input to generate a text string using the customized speech-to-text model; and using the text string as input for an intent inference model of the digital assistant.
In some embodiments, generating the one or more alternative expressions of the first speech input in the first language further includes: identifying a second speech input previously provided by a native speaker of the first language, wherein the second speech input had been associated with a respective user intent that is identical to the inferred user intent of the first speech input, and wherein a task flow executed for the respective user intent had been satisfactory to said native speaker; and utilizing the second speech input as one of the alternative expressions of the first speech input. The speech input previously provided by the native speakers are good source of example expressions showing customary usage of language and vocabulary in a particular region.
In some embodiments, providing the feedback to the user introducing the alternative expressions as a more preferred input to express the inferred user intent further includes: providing, in a second language, an explanation of a difference between a first alternative expression and the first speech input, wherein the second language is a primary language associated with the user, and the first language is a secondary language associated with the user.
In some embodiments, the method further includes: receiving a second speech input in the first language from the user, the second speech input utilizing at least one of the alternative expressions; determining whether the second speech input is a satisfactory vocal utterance of the at least one alternative expression; and upon determining that the second speech input is a satisfactory vocal utterance of the at least one alternative expression, executing a task flow to fulfill the inferred user intent.
In some embodiments, the method further includes: providing, in a second language, a paraphrase of the first speech input based on the inferred user intent to confirm the correctness of the inferred user intent, wherein the digital assistant generates the alternative expressions and provides the feedback after receiving user confirmation that the inferred user intent is the correct user intent.
In some embodiments, inferring the user intent based on at least the first speech input in the first language further includes inferring the user intent further based on a current context associated with the user.
In some embodiments, the current context associated with the user includes at least a current location of the user.
In some embodiments, the current context associated with the user includes at least a current time at which the first speech input was received.
In some embodiments, the current context associated with the user includes at least a type of place that is located at the user's current location.
In some embodiments, the current context associated with the user includes at least a correlation between a schedule item of the user and the current location.
In some embodiments, the current context associated with the user includes at least a correlation between a schedule item of the user and the current time.
In some embodiments, the current context associated with the user includes at least a current transportation mode of the user.
In some embodiments, the current context associated with the user includes at least a correlation between a directions request entered by the user and the user's current location.
In some embodiments, the method further includes storing the one or more alternative expressions for future review by the user.
In some embodiments, the method further implements features of any combination of the methods described above and in the remainder of this specification.
Accordingly, some embodiments provide a method for operating a digital assistant, the method including, at a device including one or more processors and memory storing one or more programs: receiving, from a user, a first speech input spoken in a first language; inferring a user intent based on at least the first speech input; based on the inferred user intent, generating one or more alternative expressions of the first speech input in a second language; and providing feedback to the user introducing the alternative expressions as a means to accomplish the inferred user intent when the user speaks at least one of the one or more alternative expressions to another user who understands the second language. In these embodiments, the digital assistant does not provide information or perform the task requested by the user, instead, the digital assistant teaches what the user needs to say to another person to obtain the information and/or to get the task accomplished. This is useful when the digital assistant can only correctly infer the user's intent, but does not have sufficient capabilities to accomplish the task for the user. Instead, the digital assistant teaches the user the correct foreign language expressions, such that the user can solicit and employ the help of another person who does not speak the native language of the user.
In some embodiments, the first language is a primary language associated with the user, and the second language is a primary language associated with a geographic area in which the user is currently located.
In some embodiments, the first language is a primary language associated with the user, and the second language is a secondary language associated with the user.
In some embodiments, the second language is different from the first language and at least one of the alternative expressions is not a translation of the first speech input from the first language to the second language.
In some embodiments, the digital assistant generates the alternative expressions and provides the feedback in a foreign language assistance mode in response to user selection of the foreign language assistance mode.
In some embodiments, the digital assistant initiates a foreign language assistance mode in response to detecting that the user's current location is outside of a geographic area for which the first language is a primary language, and wherein the digital assistant generates the alternative expressions and provides the feedback in the foreign language assistance mode.
In some embodiments, the digital assistant initiates a foreign language assistance mode in response to detecting that the user's current location is outside of a geographic area for which the first language is a primary language, and that the digital assistant is not able to fulfill the inferred user intent.
In some embodiments, the method further includes: in the feedback provided to the user, presenting, in the first language, a name of the second language as a respective language of the one or more alternative expressions.
In some embodiments, the method further includes: providing a practice session for the user to vocally practice at least one of the one or more alternative expressions; and during the practice session: receiving a second speech input from the user speaking at least one of the one or more alternative expressions; determining whether the second speech input is a satisfactory vocal utterance of the at least one alternative expressions; and upon determining that the second speech input is a satisfactory vocal utterance of the at least one alternative expressions, providing an output to the user indicating that the second speech input is satisfactory.
In some embodiments, the method further includes: during the practice session: providing, to the user, a sample vocal utterance for at least one of the one or more alternative expressions.
In some embodiments, the method further includes during the practice session: receiving a third speech input from the user speaking at least one of the one or more alternative expressions; detecting an error in the third speech input based on a difference between the third speech input and a standard vocal utterance of the at least one alternative expressions; and providing a sample vocal utterance to the user one or more times, the sample vocal utterance tailored for correcting the error in the third speech input.
In some embodiments, the first language is a first dialect of a respective language associated with the user, and the second language is a second dialect of the respective language, and wherein the second dialect is different from the first dialect and is associated with a respective geographic area in which the user is currently located.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that changes a pronunciation of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that changes a grammatical usage of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes at least a respective alternative expression that replaces at least one word in the first speech input.
In some embodiments, the respective alternative expression that replaces at least one word or expression in the first speech input is a local slang for the at least one word or expression in the geographic area in which the user is currently located.
In some embodiments, the digital assistant generates the alternative expressions and provides the feedback in a foreign language assistance mode, and the method further includes: while in the foreign language assistance mode: receiving input from the user for entering a live session for the user to utilize at least one of the alternative expressions to accomplish the inferred user intent; and providing the live session for the user; and during the live session: listening for the user speaking the at least one of the alternative expression to a second user; listening for a verbal response from the second user; based on the verbal response received from the second user, determining that additional foreign language assistance is needed by the user; and providing one or more speech outputs in the second language to assist the user in accomplishing the inferred user intent.
In some embodiments, the method further includes: providing, to the user, a textual transcript of a verbal exchange between the digital assistant and the second user in a user interface displayed on the device.
In some embodiments, the method further includes: providing, to the user, a translation of the textual transcript from the second language to the first language in the user interface displayed on the device.
In some embodiments, the method further includes storing a transcript of a user session conducted in the foreign language assistance mode for future review by the user.
In some embodiments, the digital assistant generates a different set of alternative expressions for the inferred user intent depending on a respective current context associated with the user.
In some embodiments, the current context associated with the user includes a current location of the user.
In some embodiments, the current context associated with the user includes a current time at which the first speech input was received.
In some embodiments, the current context associated with the user includes a type of place that is located at the user's current location.
In some embodiments, the current context associated with the user includes a correlation between a schedule item of the user and the current location.
In some embodiments, the current context associated with the user includes a correlation between a schedule item of the user and the current time.
In some embodiments, the current context associated with the user includes a current transportation mode of the user.
In some embodiments, the current context associated with the user includes a correlation between a directions request entered by the user and the user's current location.
In some embodiments, the method further implements features of any combination of the methods described above and in the remainder of this specification.
Accordingly, some embodiments provide a method for operating a digital assistant, the method including, at a device including one or more processors and memory storing one or more programs: (1) during a first interaction with a user: receiving a first speech input from the user while the user is located in a first geographic area; inferring a first user intent based on the first speech input; providing a first paraphrase of the first speech input based on the inferred first user intent; and optionally executing a respective task flow to accomplish the inferred first user intent; (2) during a second interaction with the user: receiving a second speech input from the user while the user is located in a second geographic, the second speech input being substantially identical to the first speech input; inferring a second user intent based on the second speech input, the inferred second user intent being identical to the inferred first user intent; determining that a location change from the first geographic area to the second geographic area is associated with a change in language or locale-specific vocabulary for at least one word or expression in the second speech input; in response to said determination, providing a second paraphrase based on the second inferred user intent, wherein the second paraphrase is different from the first paraphrase based on the change in language or vocabulary; and optionally executing the respective task flow to accomplish the inferred second user intent.
In some embodiments, the first geographic area and the second geographic area are both associated with a primary language of the user.
In some embodiments, the change in locale-specific vocabulary includes use of a respective local slang in the second geographic area for the at least one word or expression in the second speech input, and wherein the second paraphrase utilizes the respective local slang.
In some embodiments, the method further includes: receiving user input to start a learning session regarding the respective local slang provided in the second paraphrase; and in response to receiving the user input, providing an explanation of the usage of the respective local slang in the second geographic area.
In some embodiments, the change in language includes use of a respective local accent in the second geographic area for the at least one word or expression in the second speech input, and wherein the second paraphrase utilizes the respective local accent.
In some embodiments, the method further includes: receiving user input to start a learning session regarding the respective local accent provided in the second paraphrase; and in response to receiving the user input, providing one or more additional examples of the usage of the respective local accent in the second geographic area.
In some embodiments, the method further implements features of any combination of the methods described above and in the remainder of this specification.
Accordingly, some embodiments provide a method for operating a digital assistant, the method including, at a device including one or more processors and memory storing one or more programs: evaluating a present context associated with a user; identifying a respective foreign language training scenario associate with the present context; and providing a foreign language training session for the user, the foreign language training session containing one or more foreign language exercises tailored for the current context. In some embodiments, the digital assistant keeps track of the present context associated with the user based on various sensors (e.g., GPS, temperature sensors, light sensors, accelerometers, compass, etc.) and events occurring on the device (e.g., phone calls, email communications, notifications generated, alerts by schedule items, searches performed, and directions request fulfilled, etc.).
In some embodiments, the method further includes: automatically, without user intervention, selecting a respective language for the one or more foreign language exercises based on a primary language associated with a geographic area in which the user is currently located; and generating the one or more foreign language exercises in the automatically selected language.
In some embodiments, the method further includes: receiving user input selecting a respective language for the one or more foreign language exercises; and generating the one or more foreign language exercises in the user-selected language.
In some embodiments, the present context associated with the user includes the user's presence inside a store located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with shopping in the store.
In some embodiments, the present context associated with the user includes the user's presence in proximity to a terminal of public transportation located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with use of the public transportation.
In some embodiments, the present context associated with the user includes the user's presence inside a dining facility located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with dining at the dining facility.
In some embodiments, the present context associated with the user includes the user's presence inside a lodging facility located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with lodging at the lodging facility.
In some embodiments, the present context associated with the user includes the user's presence inside a public transport vehicle moving toward a destination for which the user has recently requested directions and the destination is located in a geographic area in which a respective foreign language is a primary language, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with visiting to said destination.
In some embodiments, the present context associated with the user includes the user's presence inside a healthcare facility, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with obtaining healthcare services at the healthcare facility.
In some embodiments, the present context associated with the user includes the user's presence inside a business premise offering beverage services, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with ordering beverages at the business premise.
In some embodiments, the method further includes presenting images associated with vocabulary used in the foreign language exercises.
In some embodiments, the method further implements features of any combination of the methods described above and in the remainder of this specification.
The above embodiments, and other embodiments described in this specification may help realize one or more of the following advantages.
In some embodiments, the digital assistant provides alternative expressions for a user input based on the user intent inferred from the user input. The alternative expressions may be more grammatically correct, have better pronunciation, and/or are more customary to the geographic region in which the user is currently located. This is helpful for non-native speakers visiting a foreign country (e.g., overseas students, business travelers, tourists, etc.), to learn and remember the local language and usage in context.
In addition, in some embodiments, the alternative expressions are not necessarily a direct translation of the user input, but are based on the user intent inferred from the user input. Alternative expressions generated based on intent is more tolerant of informalities, errors, and missing information in the user input than results produced by direct translation. Thus, many limitations of direct translation may be avoided.
In addition, in some embodiments, intent inference is contextual, and can leverage many sources of information that the digital assistant has about the user. Thus, more appropriate expressions to accomplish the true intent of the user can be provided by the digital assistant.
In some embodiments, the better expressions may be introduced to the user in the user's native language, such that the user can better understand the subtle differences between the alternative expressions provided by the digital assistant and his/her direct input.
In some embodiments, the digital assistant teaches foreign language alternative expressions to a user after the user has expressed his/her intent in his/her native language. This is useful when the user wishes to speak to a real person about his/her needs in the foreign language (e.g., when the real person only understands the foreign language), and wants the digital assistant to teach him/her the correct expressions in that foreign language. Since the capabilities of a digital assistant are sometimes limited and some assistance is better provided by a real person. The ability to let the user to express his/her needs in his/her native language to the digital assistant, and learn the correct foreign language expressions needed to be said to a real person, can greatly expand the helpfulness of the digital assistant in many scenarios (e.g., when the user is traveling abroad).
In some embodiments, after teaching the foreign language expressions to the user, the digital assistant can continue to listen to the conversation between the user and the third-party native speaker, and provides additional language assistance when needed.
In some embodiments, the digital assistant service provider already supports inputs and responses in multiple languages, and implementing foreign language or mixed language assistance to a user can effectively leverage the existing natural language processing and intent processing capabilities of the digital assistant.
In some embodiments, the digital assistant teaches the user about a locale-specific accent or slang when the user is found at that location. For example, if the user provides an input in one location, and the input is completely congruent with the local language usage, the digital assistant performs the task without providing any language training. However, when the user provides the same input at a different location, and the user input differs from the local language usage in one or more aspects, the digital assistant optionally provides some “in-context language training” about those differences, in addition to performing the task. This is helpful for both native speakers of a language and non-native speakers for a language, since there are many regional variations (e.g., expressions, word usage, accents, slangs, etc.) even for the same language.
In some embodiments, in order to be less intrusive in the training process, the digital assistant provides the differences between the user's input and the locale-specific expressions in a paraphrase of the user input, such that the user can hear the differences, and learn the new language information without expending too much time and efforts. In some embodiments, additional information about those locale-specific language differences may be provided to the user upon user request.
In some embodiments, the digital assistant keeps track of the current context associated with the user, and provides foreign language training opportunities to the user based on the user's current context. For example, training exercises related to shopping can be provided to the user, when the digital assistant detects that the user is inside a store. More specific training exercises and vocabulary may be provided to the user depending on the type of store that the user is in. Sometimes, in addition to the type of places that the user is current visiting, previous directions requests, the user's calendar items, and searches performed by the user can also be used to determine the current context, which is then used to generate relevant foreign language training exercises for the user. These contextually relevant foreign language exercises can help motivate the user's learning, and the surrounding environment also provides additional opportunities and visual cues to help the user practice and memorize the content of the foreign language exercises.
The details of one or more embodiments of the subject matter described in this specification are set forth in the accompanying drawings and the description below. Other features, aspects, and advantages of the subject matter will become apparent from the description, the drawings, and the claims.
Like reference numerals refer to corresponding parts throughout the drawings.
Specifically, a digital assistant is capable of accepting a user request at least partially in the form of a natural language command, request, statement, narrative, and/or inquiry. Typically, the user request seeks either an informational answer or performance of a task by the digital assistant. A satisfactory response to the user request is either provision of the requested informational answer, performance of the requested task, or a combination of the two. For example, a user may ask the digital assistant a question, such as “Where am I right now?” Based on the user's current location, the digital assistant may answer, “You are in Central Park.” The user may also request the performance of a task, for example, “Please remind me to call mom at 4 pm today.” In response, the digital assistant may acknowledge the request and then creates an appropriate reminder item in the user's electronic schedule. During performance of a requested task, the digital assistant sometimes interacts with the user in a continuous dialogue involving multiple exchanges of information over an extended period of time. There are numerous other ways of interacting with a digital assistant to request information or performance of various tasks. In addition to providing verbal responses and taking programmed actions, the digital assistant also provides responses in other visual or audio forms, e.g., as text, alerts, music, videos, animations, etc. In some embodiments, the digital assistant accepts input in more than one language, and provides responses in the language of the input, the user's primary language, a user's selected language, and/or a mixture of multiple languages.
An example of a digital assistant is described in Applicant's U.S. Utility application Ser. No. 12/987,982 for “Intelligent Automated Assistant,” filed Jan. 10, 2011, the entire disclosure of which is incorporated herein by reference.
As shown in
In some embodiments, the DA server 106 includes a client-facing I/O interface 112, one or more processing modules 114, data and models 116, and an I/O interface to external services 118. The client-facing I/O interface facilitates the client-facing input and output processing for the digital assistant server 106. The one or more processing modules 114 utilize the data and models 116 to determine the user's intent based on natural language input and perform task execution based on inferred user intent. In some embodiments, the DA-server 106 communicates with external services 120 through the network(s) 110 for task completion or information acquisition. The I/O interface to external services 118 facilitates such communications.
Examples of the user device 104 include, but are not limited to, a handheld computer, a personal digital assistant (PDA), a tablet computer, a laptop computer, a desktop computer, a cellular telephone, a smart phone, an enhanced general packet radio service (EGPRS) mobile phone, a media player, a navigation device, a game console, a television, a remote control, or a combination of any two or more of these data processing devices or other data processing devices. More details on the user device 104 are provided in reference to an exemplary user device 104 shown in
Examples of the communication network(s) 110 include local area networks (“LAN”) and wide area networks (“WAN”), e.g., the Internet. The communication network(s) 110 are, optionally, implemented using any known network protocol, including various wired or wireless protocols, such as e.g., Ethernet, Universal Serial Bus (USB), FIREWIRE, Global System for Mobile Communications (GSM), Enhanced Data GSM Environment (EDGE), code division multiple access (CDMA), time division multiple access (TDMA), Bluetooth, Wi-Fi, voice over Internet Protocol (VoIP), Wi-MAX, or any other suitable communication protocol.
The server system 108 is implemented on one or more standalone data processing apparatus or a distributed network of computers. In some embodiments, the server system 108 also employs various virtual devices and/or services of third party service providers (e.g., third-party cloud service providers) to provide the underlying computing resources and/or infrastructure resources of the server system 108.
Although the digital assistant shown in
For example, a motion sensor 210, a light sensor 212, and a proximity sensor 214 are coupled to the peripherals interface 206 to facilitate orientation, light, and proximity sensing functions. One or more other sensors 216, such as a positioning system (e.g., GPS receiver), a temperature sensor, a biometric sensor, a gyro, a compass, an accelerometer, and the like, are also connected to the peripherals interface 206, to facilitate related functionalities.
In some embodiments, a camera subsystem 220 and an optical sensor 222 are utilized to facilitate camera functions, such as taking photographs and recording video clips. Communication functions are facilitated through one or more wired and/or wireless communication subsystems 224, which can include various communication ports, radio frequency receivers and transmitters, and/or optical (e.g., infrared) receivers and transmitters. An audio subsystem 226 is coupled to speakers 228 and a microphone 230 to facilitate voice-enabled functions, such as voice recognition, voice replication, digital recording, and telephony functions.
In some embodiments, an I/O subsystem 240 is also coupled to the peripheral interface 206. The I/O subsystem 240 includes a touch screen controller 242 and/or other input controller(s) 244. The touch-screen controller 242 is coupled to a touch screen 246. The touch screen 246 and the touch screen controller 242 can, for example, detect contact and movement or break thereof using any of a plurality of touch sensitivity technologies, such as capacitive, resistive, infrared, surface acoustic wave technologies, proximity sensor arrays, and the like. The other input controller(s) 244 can be coupled to other input/control devices 248, such as one or more buttons, rocker switches, thumb-wheel, infrared port, USB port, and/or a pointer device such as a stylus.
In some embodiments, the memory interface 202 is coupled to memory 250. The memory 250 can include high-speed random access memory and/or non-volatile memory, such as one or more magnetic disk storage devices, one or more optical storage devices, and/or flash memory (e.g., NAND, NOR).
In some embodiments, the memory 250 stores an operating system 252, a communication module 254, a graphical user interface module 256, a sensor processing module 258, a phone module 260, and applications 262. The operating system 252 includes instructions for handling basic system services and for performing hardware dependent tasks. The communication module 254 facilitates communicating with one or more additional devices, one or more computers and/or one or more services. The graphical user interface module 256 facilitates graphic user interface processing. The sensor processing module 258 facilitates sensor-related processing and functions. The phone module 260 facilitates phone-related processes and functions. The application module 262 facilitates various functionalities of user applications, such as electronic-messaging, web browsing, media processing, Navigation, imaging and/or other processes and functions.
As described in this specification, the memory 250 also stores client-side digital assistant instructions (e.g., in a digital assistant client module 264) and various user data 266 (e.g., user-specific vocabulary data, preference data, and/or other data such as the user's electronic address book, to-do lists, shopping lists, etc.) to provide the client-side functionalities of the digital assistant.
In various embodiments, the digital assistant client module 264 is capable of accepting voice input (e.g., speech input), text input, touch input, and/or gestural input through various user interfaces (e.g., the I/O subsystem 244) of the user device 104. The digital assistant client module 264 is also capable of providing output in audio (e.g., speech output), visual, and/or tactile forms. For example, output can be provided as voice, sound, alerts, text messages, menus, graphics, videos, animations, vibrations, and/or combinations of two or more of the above. During operation, the digital assistant client module 264 communicates with the digital assistant server using the communication subsystems 224.
In some embodiments, the digital assistant client module 264 utilizes the various sensors, subsystems peripheral devices to gather additional information from the surrounding environment of the user device 104 to establish a context associated with a user, the current user interaction, and/or the current user input. In some embodiments, the digital assistant client module 264 provides the context information or a subset thereof with the user input to the digital assistant server to help infer the user's intent. In some embodiments, the digital assistant also uses the context information to determine how to prepare and delivery outputs to the user.
In some embodiments, the context information that accompanies the user input includes sensor information, e.g., lighting, ambient noise, ambient temperature, images or videos of the surrounding environment, etc. in some embodiments, the context information also includes the physical state of the device, e.g., device orientation, device location, device temperature, power level, speed, acceleration, motion patterns, cellular signals strength, etc. In some embodiments, information related to the software state of the user device 106, e.g., running processes, installed programs, past and present network activities, background services, error logs, resources usage, etc., of the user device 104 are provided to the digital assistant server as context information associated with a user input.
In some embodiments, the DA client module 264 selectively provides information (e.g., user data 266) stored on the user device 104 in response to requests from the digital assistant server. In some embodiments, the digital assistant client module 264 also elicits additional input from the user via a natural language dialogue or other user interfaces upon request by the digital assistant server 106. The digital assistant client module 264 passes the additional input to the digital assistant server 106 to help the digital assistant server 106 in intent inference and/or fulfillment of the user's intent expressed in the user request.
In various embodiments, the memory 250 includes additional instructions or fewer instructions. Furthermore, various functions of the user device 104 may be implemented in hardware and/or in firmware, including in one or more signal processing and/or application specific integrated circuits.
The digital assistant system 300 includes memory 302, one or more processors 304, an input/output (I/O) interface 306, and a network communications interface 308. These components communicate with one another over one or more communication buses or signal Lines 310.
In some embodiments, the memory 302 includes a non-transitory computer readable medium, such as high-speed random access memory and/or a non-volatile computer readable storage medium (e.g., one or more magnetic disk storage devices, flash memory devices, or other non-volatile solid-state memory devices).
In some embodiments, the I/O interface 306 couples input/output devices 316 of the digital assistant system 300, such as displays, a keyboards, touch screens, and microphones, to the user interface module 322. The I/O interface 306, in conjunction with the user interface module 322, receive user inputs (e.g., voice input, keyboard inputs, touch inputs, etc.) and process them accordingly. In some embodiments, e.g., when the digital assistant is implemented on a standalone user device, the digital assistant system 300 includes any of the components and I/O and communication interfaces described with respect to the user device 104 in
In some embodiments, the network communications interface 308 includes wired communication port(s) 312 and/or wireless transmission and reception circuitry 314. The wired communication port(s) receive and send communication signals via one or more wired interfaces, e.g., Ethernet, Universal Serial Bus (USB), FIREWIRE, etc. The wireless circuitry 314 receives and sends RF signals and/or optical signals from/to communications networks and other communications devices. The wireless communications, optionally, use any of a plurality of communications standards, protocols and technologies, such as GSM, EDGE, CDMA, TDMA, Bluetooth, Wi-Fi, VoIP, Wi-MAX, or any other suitable communication protocol. The network communications interface 308 enables communication between the digital assistant system 300 with networks, such as the Internet, an intranet and/or a wireless network, such as a cellular telephone network, a wireless local area network (LAN) and/or a metropolitan area network (MAN), and other devices.
In some embodiments, memory 302, or the computer readable storage media of memory 302, stores programs, modules, instructions, and data structures including all or a subset of: an operating system 318, a communications module 320, a user interface module 322, one or more applications 324, and a digital assistant module 326. The one or more processors 304 execute these programs, modules, and instructions, and reads/writes from/to the data structures.
The operating system 318 (e.g., Darwin, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks) includes various software components and/or drivers for controlling and managing general system tasks (e.g., memory management, storage device control, power management, etc.) and facilitates communications between various hardware, firmware, and software components.
The communications module 320 facilitates communications between the digital assistant system 300 with other devices over the network communications interface 308. For example, the communication module 320, optionally, communicates with the communication interface 254 of the device 104 shown in
The user interface module 322 receives commands and/or inputs from a user via the I/O interface 306 (e.g., from a keyboard, touch screen, pointing device, controller, and/or microphone), and generates user interface objects on a display. The user interface module 322 also prepares and delivers outputs (e.g., speech, sound, animation, text, icons, vibrations, haptic feedback, and light, etc.) to the user via the I/O interface 306 (e.g., through displays, audio channels, speakers, and touch-pads, etc.).
The applications 324 include programs and/or modules that are configured to be executed by the one or more processors 304. For example, if the digital assistant system is implemented on a standalone user device, the applications 324, optionally, include user applications, such as games, a calendar application, a navigation application, or an email application. If the digital assistant system 300 is implemented on a server farm, the applications 324, optionally, include resource management applications, diagnostic applications, or scheduling applications, for example.
The memory 302 also stores the digital assistant module (or the server portion of a digital assistant) 326. In some embodiments, the digital assistant module 326 includes the following sub-modules, or a subset or superset thereof: an input/output processing module 328, a speech-to-text (STT) processing module 330, a natural language processing module 332, a dialogue flow processing module 334, a task flow processing module 336, a service processing module 338, and a user training module 340. Each of these modules has access to one or more of the following data and models of the digital assistant 326, or a subset or superset thereof: ontology 360, vocabulary index 344, user data 348, task flow models 354, service models 356, and user training data 358.
In some embodiments, using the processing modules, data, and models implemented in the digital assistant module 326, the digital assistant performs at least some of the following: identifying a user's intent expressed in a natural language input received from the user; actively eliciting and obtaining information needed to fully infer the user's intent (e.g., by disambiguating words, names, intentions, etc.); determining the task flow for fulfilling the inferred intent; and executing the task flow to fulfill the inferred intent. In this specifications, more details regarding the user training module 340 and its use of the user training data 358 are provided later in
In some embodiments, as shown in
The speech-to-text processing module 330 receives Speech input (e.g., a user utterance captured in a voice recording) through the I/O processing module 328. In some embodiments, the speech-to-text processing module 330 uses various acoustic and language models to recognize the speech input as a sequence of phonemes, and ultimately, a sequence of words or tokens written in one or more languages. The speech-to-text processing module 330 can be implemented using any suitable speech recognition techniques, acoustic models, and language models, such as Hidden Markov Models, Dynamic Time Warping (DTW)-based speech recognition, and other statistical and/or analytical techniques. In some embodiments, the speech-to-text processing can be performed at least partially by a third party service or on the user's device. In some embodiments, the speech-to-text processing module 330 handles input in multiple languages and have locale-specific acoustic models and language models tailored for each language and locale-specific inputs. In some embodiments, the speech-to-text processing module 330 includes multiple acoustic and language models for each single language to accommodate regional and other variations. Once the speech-to-text processing module 330 obtains the result of the speech-to-text processing, e.g., a sequence of words or tokens, it passes the result to the natural language processing module 332 for intent inference.
More details on the speech-to-text processing are described in U.S. Utility application Ser. No. 13/236,942 for “Consolidating Speech Recognition Results,” filed on Sep. 20, 2011, the entire disclosure of which is incorporated herein by reference.
The natural language processing module 332 (“natural language processor”) of the digital assistant takes the sequence of words or tokens (“token sequence”) generated by the speech-to-text processing module 330, and attempts to associate the token sequence with one or more “actionable intents” recognized by the digital assistant. An “actionable intent” represents a task that can be performed by the digital assistant, and has an associated task flow implemented in the task flow models 354. The associated task flow is a series of programmed actions and steps that the digital assistant takes in order to perform the task. The scope of a digital assistant's capabilities is dependent on the number and variety of task flows that have been implemented and stored in the task flow models 354, or in other words, on the number and variety of “actionable intents” that the digital assistant recognizes. The effectiveness of the digital assistant, however, is also dependent on the assistant's ability to infer the correct “actionable intent(s)” from the user request expressed in natural language.
In some embodiments, the natural language processing module 332 handles input in multiple languages and have locale-specific vocabulary and language usage models tailored for each language and locale-specific inputs. In some embodiments, the natural language processing module 332 includes multiple locale-specific vocabulary and language usage models for each single language to accommodate regional and other variations.
In some embodiments, in addition to the sequence of words or tokens obtained from the speech-to-text processing module 330, the natural language processor 332 also receives context information associated with the user request, e.g., from the I/O processing module 328. The natural language processor 332, optionally, uses the context information to clarify, supplement, and/or further define the information contained in the token sequence received from the speech-to-text processing module 330. The context information includes, for example, user preferences, hardware and/or software states of the user device, sensor information collected before, during, or shortly after the user request, prior interactions (e.g., dialogue) between the digital assistant and the user, and the like.
In some embodiments, the natural language processing is based on ontology 360. The ontology 360 is a hierarchical structure containing many nodes, each node representing either an “actionable intent” or a “property” relevant to one or more of the “actionable intents” or other “properties”. As noted above, an “actionable intent” represents a task that the digital assistant is capable of performing, i.e., it is “actionable” or can be acted on. A “property” represents a parameter associated with an actionable intent, a domain concept or entity, or a sub-aspect of another property. A linkage between an actionable intent node and a property node in the ontology 360 defines how a parameter represented by the property node pertains to the task represented by the actionable intent node.
In some embodiments, the ontology 360 is made up of actionable intent nodes and property nodes. Within the ontology 360, each actionable intent node is linked to one or more property nodes either directly or through one or more intermediate property nodes. Similarly, each property node is linked to one or more actionable intent nodes either directly or through one or more intermediate property nodes. For example, as shown in
An actionable intent node, along with its linked concept nodes, may be described as a “domain.” In the present discussion, each domain is associated with a respective actionable intent, and refers to the group of nodes (and the relationships therebetween) associated with the particular actionable intent. For example, the ontology 360 shown in
While
In some embodiments, the ontology 360 includes all the domains (and hence actionable intents) that the digital assistant is capable of understanding and acting upon. In some embodiments, the ontology 360 is optionally modified, such as by adding or removing entire domains or nodes, or by modifying relationships between the nodes within the ontology 360.
In some embodiments, nodes associated with multiple related actionable intents are optionally clustered under a “super domain” in the ontology 360. For example, a “travel” super-domain optionally includes a cluster of property nodes and actionable intent nodes related to travels. The actionable intent nodes related to travels optionally includes “airline reservation,” “hotel reservation,” “car rental,” “get directions,” “find points of interest,” and so on. The actionable intent nodes under the same super domain (e.g., the “travels” super domain) sometimes have many property nodes in common. For example, the actionable intent nodes for “airline reservation,” “hotel reservation,” “car rental,” “get directions,” “find points of interest” sometimes share one or more of the property nodes “start location,” “destination,” “departure date/time,” “arrival date/time,” and “party size.”
In some embodiments, each node in the ontology 360 is associated with a set of words and/or phrases that are relevant to the property or actionable intent represented by the node. The respective set of words and/or phrases associated with each node is the so-called “vocabulary” associated with the node. The respective set of words and/or phrases associated with each node can be stored in the vocabulary index 344 in association with the property or actionable intent represented by the node. For example, returning to
The natural language processor 332 receives the token sequence (e.g., a text string) from the speech-to-text processing module 330, and determines what nodes are implicated by the words in the token sequence. In some embodiments, if a word or phrase in the token sequence is found to be associated with one or more nodes in the ontology 360 (via the vocabulary index 344), the word or phrase will “trigger” or “activate” those nodes. Based on the quantity and/or relative importance of the activated nodes, the natural language processor 332 will select one of the actionable intents as the task that the user intended the digital assistant to perform. In some embodiments, the domain that has the most “triggered” nodes is selected. In some embodiments, the domain having the highest confidence value (e.g., based on the relative importance of its various triggered nodes) is selected. In some embodiments, the domain is selected based on a combination of the number and the importance of the triggered nodes. In some embodiments, additional factors are considered in selecting the node as well, such as whether the digital assistant has previously correctly interpreted a similar request from a user.
In some embodiments, the digital assistant also stores names of specific entities in the vocabulary index 344, so that when one of these names is detected in the user request, the natural language processor 332 will be able to recognize that the name refers to a specific instance of a property or sub-property in the ontology. In some embodiments, the names of specific entities are names of businesses, restaurants, people, movies, and the like. In some embodiments, the digital assistant searches and identifies specific entity names from other data sources, such as the user's address book, a movies database, a musicians database, and/or a restaurant database. In some embodiments, when the natural language processor 332 identifies that a word in the token sequence is a name of a specific entity (such as a name in the user's address book), that word is given additional significance in selecting the actionable intent within the ontology for the user request.
For example, when the words “Mr. Santo” are recognized from the user request, and the last name “Santo” is found in the vocabulary index 344 as one of the contacts in the user's contact list, then it is likely that the user request corresponds to a “send a message” or “initiate a phone call” domain. For another example, when the words “ABC Café” are found in the user request, and the term “ABC Café” is found in the vocabulary index 344 as the name of a particular restaurant in the user's city, then it is likely that the user request corresponds to a “restaurant reservation” domain.
User data 348 includes user-specific information, such as user-specific vocabulary, user preferences, user address, user's default and secondary languages, user's contact list, and other short-term or long-term information for each user. In some embodiments, the natural language processor 332 uses the user-specific information to supplement the information contained in the user input to further define the user intent. For example, for a user request “invite my friends to my birthday party,” the natural language processor 332 is able to access user data 348 to determine who the “friends” are and when and where the “birthday party” would be held, rather than requiring the user to provide such information explicitly in his/her request.
Other details of searching an ontology based on a token string is described in U.S. Utility application Ser. No. 12/341,743 for “Method and Apparatus for Searching Using An Active Ontology,” filed Dec. 22, 2008, the entire disclosure of which is incorporated herein by reference.
In some embodiments, once the natural language processor 332 identifies an actionable intent (or domain) based on the user request, the natural language processor 332 generates a structured query to represent the identified actionable intent. In some embodiments, the structured query includes parameters for one or more nodes within the domain for the actionable intent, and at least some of the parameters are populated with the specific information and requirements specified in the user request. For example, the user may say “Make me a dinner reservation at a sushi place at seven o′clock.” In this case, the natural language processor 332 may be able to correctly identify the actionable intent to be “restaurant reservation” based on the user input. According to the ontology, a structured query for a “restaurant reservation” domain optionally includes parameters such as {Cuisine}, {Time}, {Date}, {Party Size}, and the like. In some embodiments, based on the information contained in the user's utterance, the natural language processor 332 generates a partial structured query for the restaurant reservation domain, where the partial structured query includes the parameters {Cuisine=“Sushi”} and {Time=“7 pm”}. However, in this example, the user's utterance contains insufficient information to complete the structured query associated with the domain. Therefore, other necessary parameters such as {Party Size} and {Date} are not specified in the structured query based on the information currently available. In some embodiments, the natural language processor 332 populates some parameters of the structured query with received context information. For example, in some embodiments, if the user requested a sushi restaurant “near me,” the natural language processor 332 populates a {location} parameter in the structured query with GPS coordinates from the user device 104.
In some embodiments, the natural language processor 332 passes the structured query (including any completed parameters) to the task flow processing module 336 (“task flow processor”). The task flow processor 336 is configured to receive the structured query from the natural language processor 332, complete the structured query, if necessary, and perform the actions required to “complete” the user's ultimate request. In some embodiments, the various procedures necessary to complete these tasks are provided in task flow models 354. In some embodiments, the task flow models include procedures for obtaining additional information from the user, and task flows for performing actions associated with the actionable intent.
As described above, in order to complete a structured query, the task flow processor 336 sometimes needs to initiate additional dialogue with the user in order to obtain additional information, and/or disambiguate potentially ambiguous utterances. When such interactions are necessary, the task flow processor 336 invokes the dialogue processing module 334 (“dialogue processor 334”) to engage in a dialogue with the user. In some embodiments, the dialogue processor 334 determines how (and/or when) to ask the user for the additional information, and receives and processes the user responses. The questions are provided to and answers are received from the users through the I/O processing module 328. In some embodiments, the dialogue processor 334 presents dialogue output to the user via audio and/or visual output, and receives input from the user via spoken or physical (e.g., clicking) responses. Continuing with the example above, when the task flow processor 336 invokes the dialogue flow processor 334 to determine the “party size” and “date” information for the structured query associated with the domain “restaurant reservation,” the dialogue flow processor 335 generates questions such as “For how many people?” and “On which day?” to pass to the user. Once answers are received from the user, the dialogue flow processor 334 can then populate the structured query with the missing information, or pass the information to the task flow processor 336 to complete the missing information from the structured query.
In some cases, the task flow processor 336 receives a structured query that has one or more ambiguous properties, e.g., when a structured query for the “send a message” domain indicates that the intended recipient is “Bob,” and the user happens to have multiple contacts named “Bob.” In such cases, the task flow processor 336, optionally, requests that the dialogue processor 334 disambiguate this property of the structured query. In turn, the dialogue processor 334, optionally, asks the user “Which Bob?”, and displays (or reads) a list of contacts named “Bob” from which the user may choose.
Once the task flow processor 336 has completed the structured query for an actionable intent, the task flow processor 336 proceeds to perform the ultimate task associated with the actionable intent. Accordingly, the task flow processor 336 executes the steps and instructions in the task flow model according to the specific parameters contained in the structured query. For example, the task flow model for the actionable intent of “restaurant reservation”, optionally, includes steps and instructions for contacting a restaurant and actually requesting a reservation for a particular party size at a particular time. For example, using a structured query such as: {restaurant reservation, restaurant=ABC Café, date=3/12/2012, time=7 pm, party size=5}, the task flow processor 336, optionally, performs the steps of: (1) logging onto a server of the ABC Café or a restaurant reservation system such as OPENTABLE®, (2) entering the date, time, and party size information in a form on the website, (3) submitting the form, and (4) making a calendar entry for the reservation in the user's calendar.
In some embodiments, the task flow processor 336 employs the assistance of a service processing module 338 (“service processor”) to complete a task requested in the user input or to provide an informational answer requested in the user input. For example, the service processor 338 can act on behalf of the task flow processor 336 to make a phone call, set a calendar entry, invoke a map search, invoke or interact with other user applications installed on the user device, and invoke or interact with third party services (e.g. a restaurant reservation portal, a social networking website, a banking portal, etc.). In some embodiments, the protocols and application programming interfaces (API) required by each service can be specified by a respective service model among the services models 356. The service processor 338 accesses the appropriate service model for a service and generates requests for the service in accordance with the protocols and APIs required by the service according to the service model.
For example, if a restaurant has enabled an online reservation service, the restaurant can submit a service model specifying the necessary parameters for making a reservation and the APIs for communicating the values of the necessary parameter to the online reservation service. When requested by the task flow processor 336, the service processor 338 can establish a network connection with the online reservation service using the web address stored in the service model, and send the necessary parameters of the reservation (e.g., time, date, party size) to the online reservation interface in a format according to the API of the online reservation service.
in some embodiments, the natural language processor 332, dialogue processor 334, and task flow processor 336 are used collectively and iteratively to infer and define the user's intent, obtain information to further clarify and refine the user intent, and finally generate a response (i.e., an output to the user, or the completion of a task) to fulfill the user's intent.
In some embodiments, after all of the tasks needed to fulfill the user's request have been performed, the digital assistant 326 formulates a confirmation response, and sends the response back to the user through the I/O processing module 328. If the user request seeks an informational answer, the confirmation response presents the requested information to the user. In some embodiments, the digital assistant also requests the user to indicate whether the user is satisfied with the response produced by the digital assistant 326.
More details on the digital assistant can be found in the U.S. Utility application Ser. No. 12/987,982, entitled “Intelligent Automated Assistant”, filed Jan. 10, 2011, U.S. Provisional Application No. 61/493,201, entitled “Generating and Processing Data Items That Represent Tasks to Perform”, filed Jun. 3, 2011, the entire disclosures of which are incorporated herein by reference.
As described in this specification, in some embodiments, a digital assistant provides training, in particular, locale-specific language training and/or foreign language training and assistance to the user. The exemplary processes provided below may be implemented by the user training module 340, using the information stored in the user training data 358. In some embodiments, the user training data 358 includes suitable alternative expressions and vocabulary in various languages indexed by user intent, and templates for additional foreign language exercises.
In the process 400, the digital assistant receives (402), from a user, a first speech input spoken in a first language. The digital assistant infers (404) a user intent based on at least the first speech input in the first language. Based on the inferred user intent, the digital assistant generates (406) one or more alternative expressions of the first speech input in the first language. The digital assistant provides (408) feedback to the user introducing the alternative expressions as a more preferred input to express the inferred user intent than the first speech input provided by the user.
In an example scenario, the user provides a speech input in English to the digital assistant “Where can I buy a torch?” The user, being a non-native English speaker, may not be aware that the term “torch” has a different meaning in the United States than in other English-speaking countries. The digital assistant receiving the speech input is aware of the different meanings for the term “torch” in different locales (e.g., in England, “torch” refers to a type of illumination device relying on dry batteries (i.e., “flashlight” in the United States), but in the United States, “torch” refers to a type of illumination devices relying on burning an organic fuel). The digital assistant infers an intent based on the user's input, and determines that the user is more likely to be asking about the type of illumination devices using dry batteries. Based on the inferred intent and the user's current location (e.g., in the United States), the digital assistant generates at least one alternative expression for the term “torch,” such as “flashlight” and provides that alternative expression to the user. For example, in some embodiments, the digital assistant displays in a user interface names and directions to stores (e.g., hardware stores) that sell flashlights, and in addition, the digital assistant also teaches the user that the term “flashlight” is more customarily used in the United States than the term “torch.” In some embodiments, instead of teaching the user about the term “torch” directly, the digital assistant optionally provides a paraphrase of the user's speech input, where the paraphrase introduces the term “flashlight” to the user. For example, the paraphrase provided by the digital assistant can be in the form of a confirmation request “Did you mean you want to buy a “flashlight” which uses dry batteries instead of a burning fuel?” Alternatively, the digital assistant optionally says “Searching for stores nearby that sell flashlights . . . .”
In another example, when a non-native English speaker provides a speech input in English, but speaks one or more words with a heavy accent. The digital assistant can infer the user's intent based on other content in the speech input and the current context. Based on the inferred user intent, the digital assistant can generate alternative expressions that correct the accent of the particular words. For example, many proper nouns (e.g., names of international stores and businesses) are used worldwide, and the pronunciations for those proper nouns are localized in different countries and regions. When a non-native speaker speaks those proper nouns in a request to the digital assistant, they frequently use the pronunciations in their native languages, even though the rest of the request is spoken with a proper English accent. Based on the inferred user intent, the digital assistant can determine what those words are, and provide alternative expressions that have the correct pronunciation in an American or British accent. For example, when the user says “I want to fine a MaiDanglao to get a burger.” Although the pronunciation of the user is imperfect, the digital assistant infers that the user wishes to find a McDonald's to get a burger, and presents a speech output saying “OK, McDonald's is what you want to find. I will show you the directions.” The digital assistant optionally places vocal emphasis on the terms “McDonald's” and “find” to indicate to the user the proper pronunciation of these words. In some embodiments, spelling of the words “McDonald's” and “find” is shown to the user in a user interface as well.
In some embodiments, when providing the feedback to the user, the digital assistant provides (410) the feedback in a second language different from the first language, where the second language is a primary language associated with the user, and the first language is a secondary language associated with the user. For example, if the user is not a native English speaker, and provides a speech input in English, the digital assistant optionally provides the feedback in the native language of the speaker (e.g., Chinese). In other words, part of the response provided by the digital assistant is in the user's native language, while the alternative expressions are in English. Continuing with the earlier example, the digital assistant may provide a speech output “______ McDonald's ______.” The English translation of the speech output is “I think you want to find McDonald's. I found two nearby.”
In some embodiments, the one or more alternative expressions of the first speech input includes (412) at least a respective alternative expression that corrects a pronunciation of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes (414) at least a respective alternative expression that corrects a grammatical usage of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes (416) at least a respective alternative expression that replaces at least one word or phrase in the first speech input with another word or phrase.
In some embodiments, the digital assistant provide (418) at least a command mode and a foreign language training mode, where the digital assistant (1) executes a task flow to fulfill the inferred user intent in the command mode, and (2) generates the one or more alternative expressions and provides the feedback to the user in the foreign language training mode.
In some embodiments, the digital assistant concurrently provides (420) both the command mode and the foreign language training mode (e.g., in a hybrid mode), where the digital assistant executes the task flow to fulfill the inferred user intent, in addition to generating the one or more alternative expressions and providing the feedback to the user.
In some embodiments, the digital assistant receives (422) user selection of the foreign language training mode; and enables (424) the foreign language training mode in response to the user selection of the foreign language training mode.
In some embodiments, the digital assistant automatically, without user intervention, enables (426) the foreign language training mode based on a current location of the user, where a primary language associated with the current location of the user is the first language.
In some embodiments, to infer (428) the user intent based on the first speech input in the first language, the digital assistant identifies (430) a customized speech-to-text model of the first language for the user, where the customized speech-to-text model has been established based on training samples provided by native speakers of a second language of which the user is also a native speaker. The digital assistant then process (432) the first speech input to generate a text string using the customized speech-to-text model. In some embodiments, the digital assistant uses (434) the text string as input for an intent inference model of the digital assistant.
In some embodiments, to generate (436) the one or more alternative expressions of the first speech input in the first language, the digital assistant identifies (438) a second speech input previously provided by a native speaker of the first language, where the second speech input had been associated with a respective user intent that is identical to the inferred user intent of the first speech input, and where a task flow executed for the respective user intent had been satisfactory to said native speaker. The digital assistant then utilizes (440) the second speech input as one of the alternative expressions of the first speech input.
In some embodiments, to provide (442) the feedback to the user introducing the alternative expressions as a more preferred input to express the inferred user intent, the digital assistant provides (444), in a second language, an explanation of a difference between a first alternative expression and the first speech input, where the second language is a primary language associated with the user, and the first language is a secondary language associated with the user.
In some embodiments, the digital assistant receives (446) a second speech input in the first language from the user, the second speech input utilizing at least one of the alternative expressions provided to the user. In some embodiments, the digital assistant determines (448) whether the second speech input is a satisfactory vocal utterance of the at least one alternative expression. In some embodiments, upon determining that the second speech input is a satisfactory vocal utterance of the at least one alternative expression, the digital assistant executes (450) a task flow to fulfill the inferred user intent.
In some embodiments, the digital assistant provides (452), in a second language, a paraphrase of the first speech input based on the inferred user intent to confirm the correctness of the inferred user intent, where the digital assistant generates the alternative expressions and provides the feedback after receiving user confirmation that the inferred user intent is the correct user intent.
In some embodiments, inferring the user intent based on at least the first speech input in the first language further includes (454) inferring the user intent further based on a current context associated with the user.
In some embodiments, the current context associated with the user includes (456) at least a current location of the user.
In some embodiments, the current context associated with the user includes (458) at least a current time at which the first speech input was received.
In some embodiments, the current context associated with the user includes (460) at least a type of place that is located at the user's current location.
In some embodiments, the current context associated with the user includes (462) at least a correlation between a schedule item of the user and the current location.
In some embodiments, the current context associated with the user includes (464) at least a correlation between a schedule item of the user and the current time.
In some embodiments, the current context associated with the user includes (466) at least a current transportation mode of the user.
In some embodiments, the current context associated with the user includes (468) at least a correlation between a directions request entered by the user and the user's current location.
In some embodiments, the digital assistant stores (470) the one or more alternative expressions for future review by the user.
In some embodiments, the process 400 includes any combination of the features described above and in the remainder of this specification.
In the process 500, in some embodiments, the digital assistant receives (502), from a user, a first speech input spoken in a first language. The digital assistant infers (504) a user intent based on at least the first speech input. Based on the inferred user intent, the digital assistant generates (506) one or more alternative expressions of the first speech input in a second language. The digital assistant then provides (508) feedback to the user introducing the alternative expressions as a means to accomplish the inferred user intent when the user speaks at least one of the one or more alternative expressions to another user who understands the second language.
In an example scenario, the user is in a foreign country (e.g., China) and does not speak the native language (e.g., Chinese) of the region (e.g., southern China). The user can employ the assistance of his/her digital assistant, but sometimes, the capabilities of the digital assistant are not adequate in the current situation. For example, if the user is visiting a client “Mr. Santo,” and the user is already in Mr. Santo's office building. He cannot ask the digital assistant for information regarding Mr. Santo's whereabouts in the office building. Instead, the user needs to speak to a real person in the foreign language (e.g., Mandarin) that is understood in this region. In this case, the user asks the digital assistant for foreign language assistance. For example, the user may enable a foreign language assistance mode, and ask the digital assistant with a speech input in his native language (e.g., English), “I need to find Mr. Santo.” The digital assistant can correctly infer the user's intent based on the user's speech input. Instead of providing directions to the user, the digital assistant provides expressions in Chinese that would be useful for the user to enlist help of a Chinese person. For example, the digital assistant optionally provides sample speech outputs saying “______, ______ Santo ______? (meaning “Hello, which floor is Mr. Santo located?”), and/or “______, ______ Santo ______? (meaning “Sir, is Mr. Santo in?”), and/or “______, ______ Santo ______, ______? (meaning “Miss, I am looking for Mr. Santo. Is he here today?”). In some embodiments, the digital assistant provides each of these alternative expressions in Chinese, and plays back sample recordings of these alternative Chinese expressions. In some embodiments, the digital assistant also helps the user to practice the pronunciation of these expressions in Chinese, e.g., by providing phonetic spellings of these sentences.”
In another example, if the user (e.g., an English speaker) is driving in a foreign country (e.g., Taiwan), and gets a flat tire. The user may ask the digital assistant to find tire shops or towing companies on a map, but may not be able to get the services needed without speaking the local language. In some embodiments, the user enables the foreign language assistance mode, and provides a speech input explaining his/her needs. For example, the user may say to the digital assistant in English “I have a flat tire and need to call a tow truck.” The digital assistant processes the speech input, and determines that the user needs to speak to a person at a towing service. Based on the user intent, the digital assistant generates a number of expressions in the local language (e.g., Mandarin), and provides the expressions to the user. For example, in an output interface of the digital assistant, the digital assistant optionally provides the following expressions: “______, ______, ______ A ______ 10 ______ (meaning “Hi, I need towing service. I am located at the intersection of A street and No. 10 road), and/or “______, ______, ______ (meaning “Hi, my tire blew. I need a tow truck or a mechanic to come.”). In some embodiments, the digital assistant teaches the user how to say these expressions in the foreign language properly, and let the user practice a few times before letting the user call the local roadside assistance services. In some embodiments, as shown in this example, the digital assistant optionally includes additional information the assistant has about the user (e.g., the user's current location) in the expressions, even though the user him/herself may not possess this information or have included this information in his/her speech input to the digital assistant.
In some embodiments, the first language is (510) a primary language associated with the user, and the second language is a primary language associated with a geographic area in which the user is currently located.
In some embodiments, the first language is (512) a primary language associated with the user, and the second language is a secondary language associated with the user.
In some embodiments, the second language is (514) different from the first language and at least one of the alternative expressions is not a translation of the first speech input from the first language to the second language.
In some embodiments, the digital assistant generates (516) the alternative expressions and provides the feedback in a foreign language assistance mode in response to user selection of the foreign language assistance mode.
In some embodiments, the digital assistant initiates (518) a foreign language assistance mode in response to detecting that the user's current location is outside of a geographic area for which the first language is a primary language, and wherein the digital assistant generates the alternative expressions and provides the feedback in the foreign language assistance mode.
In some embodiments, the digital assistant initiates (520) a foreign language assistance mode in response to detecting that the user's current location is outside of a geographic area for which the first language is a primary language, and that the digital assistant is not able to fulfill the inferred user intent.
In some embodiments, in the feedback provided to the user, the digital assistant presents (522), in the first language, a name of the second language as a respective language of the one or more alternative expressions.
In some embodiments, the digital assistant provides (524) a practice session for the user to vocally practice at least one of the one or more alternative expressions. During the practice session (526): the digital assistant receives (528) a second speech input from the user speaking at least one of the one or more alternative expressions; determines (530) whether the second speech input is a satisfactory vocal utterance of the at least one alternative expressions; and upon determining that the second speech input is a satisfactory vocal utterance of the at least one alternative expressions, provides (532) an output to the user indicating that the second speech input is satisfactory.
In some embodiments, during the practice session, the digital assistant provides (534), to the user, a sample vocal utterance for at least one of the one or more alternative expressions.
In some embodiments, during the practice session, the digital assistant receives (536) a third speech input from the user speaking at least one of the one or more alternative expressions. In some embodiments, the digital assistant detects (538) an error in the third speech input based on a difference between the third speech input and a standard vocal utterance of the at least one alternative expressions. In some embodiments, the digital assistant provides (540) a sample vocal utterance to the user one or more times, the sample vocal utterance tailored for correcting the error in the third speech input
In some embodiments, the first language is (542) a first dialect of a respective language associated with the user, and the second language is a second dialect of the respective language, and where the second dialect is different from the first dialect and is associated with a respective geographic area in which the user is currently located.
In some embodiments, the one or more alternative expressions of the first speech input includes (544) at least a respective alternative expression that changes a pronunciation of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes (546) at least a respective alternative expression that changes a grammatical usage of at least one word in the first speech input.
In some embodiments, the one or more alternative expressions of the first speech input includes (548) at least a respective alternative expression that replaces at least one word in the first speech input.
In some embodiments, the respective alternative expression that replaces at least one word or expression in the first speech input is (550) a local slang for the at least one word or expression in the geographic area in which the user is currently located.
In some embodiments, the digital assistant generates (552) the alternative expressions and provides the feedback in a foreign language assistance mode. In some embodiments, while in the foreign language assistance mode (554), the digital assistant receives (556) input from the user for entering a live session for the user to utilize at least one of the alternative expressions to accomplish the inferred user intent. In some embodiments, the digital assistant provides (558) the live session for the user. In some embodiments, during the live session (560), the digital assistant listens (562) for the user speaking the at least one of the alternative expression to a second user. The digital assistant also listens (564) for a verbal response from the second user. Based on the verbal response received from the second user, the digital assistant determines (566) that additional foreign language assistance is needed by the user; and provides (568) one or more speech outputs in the second language to assist the user in accomplishing the inferred user intent.
In some embodiments, the digital assistant provides (570), to the user, a textual transcript of a verbal exchange between the digital assistant and the second user in a user interface displayed on the device.
In some embodiments, the digital assistant provides (572), to the user, a translation of the textual transcript from the second language to the first language in the user interface displayed on the device.
In some embodiments, the digital assistant stores (574) a transcript of a user session conducted in the foreign language assistance mode for future review by the user.
In some embodiments, the digital assistant generates (576) a different set of alternative expressions for the inferred user intent depending on a respective current context associated with the user.
In some embodiments, the current context associated with the user includes (578) a current location of the user.
In some embodiments, the current context associated with the user includes (580) a current time at which the first speech input was received.
In some embodiments, the current context associated with the user includes (582) a type of place that is located at the user's current location. Example types of places include places to shop (e.g., shopping mall, grocery stores, clothing outlets, shoe stores, electronic stores, supermarkets, etc.), places to get drinks (e.g., bars, pubs, etc.), places to get coffee or other beverages (e.g. coffee shops, juice bar, ice cream shops, tea houses, etc.), places to eat (e.g., fining dining restaurants, fast food restaurants, café, cafeteria, etc.) places to send mail (e.g., postal offices, mail boxes, commercial shipping services, etc.), places to get healthcare services (e.g., hospitals and clinics, emergency services, etc.), places to get banking services (e.g., banks, check cashing services, etc.), places to take public transportation (e.g., train stations, bus stops, airports, etc.), places to see movies (e.g., theatres, movie theatres, video stores, video rental stores, etc.), tourist sites, places to get police assistance (e.g., police station, through police dispatchers), etc.
In some embodiments, the current context associated with the user includes (584) a correlation between a schedule item of the user and the current location.
In some embodiments, the current context associated with the user includes (586) a correlation between a schedule item of the user and the current time.
In some embodiments, the current context associated with the user includes (588) a current transportation mode of the user.
In some embodiments, the current context associated with the user includes (590) a correlation between a directions request entered by the user and the user's current location.
In some embodiments, the process 500 further implements any combination of the features described above and in the remainder of this specification.
In the process 600, in some embodiments, during a first interaction with a user (602): the digital assistant receives (604) a first speech input from the user while the user is located in a first geographic area. The digital assistant infers (606) a first user intent based on the first speech input. The digital assistant provides (608) a first paraphrase of the first speech input based on the inferred first user intent. The digital assistant then executes (610) a respective task flow to accomplish the inferred first user intent. During a second interaction with the same user (612): the digital assistant receives (614) a second speech input from the user while the user is located in a second geographic, the second speech input being substantially identical to the first speech input. The digital assistant infers (616) a second user intent based on the second speech input, the inferred second user intent being identical to the inferred first user intent. The digital assistant determines (618) that a location change from the first geographic area to the second geographic area is associated with a change in language or locale-specific vocabulary for at least one word or expression in the second speech input. In response to said determination, the digital assistant provides (620) a second paraphrase based on the second inferred user intent, where the second paraphrase is different from the first paraphrase based on the change in language or vocabulary. In some embodiments, the digital assistant executes (622) the respective task flow to accomplish the inferred second user intent.
In an example scenario, during one user session, if the user says “I want to buy some pencils and erasers” while the user is in the United States, the digital assistant will infer that the user needs to find a stationery shop, and provides search results and directions to one or more stationery shops nearby. In some embodiments, to confirm the user intent, the digital assistant optionally provides a paraphrase of the user input, e.g., “Search for stores that sell erasers and pencils . . . . Here are a few stationery stores I found . . . .” When the user travels to a different location (e.g., England) where the language usage and vocabulary are somewhat different from the United States, the digital assistant optionally provides opportunities for the user to learn about the local language usage and vocabulary. For example, in another user session occurring while the user is in England, if the user says “I want to buy some erasers and pencils,” the digital assistant will infer the same user intent as before (e.g., the user needs to find a stationery shop). In addition to providing search results and directions to one or more stationery shops nearby, the digital assistant optionally provides a different paraphrase of the user input. In this paraphrase, the digital assistant can teach the user about the language difference in the U.S. and England for the term “eraser.” For example, the digital assistant optionally provides a paraphrase that says “Search for stores that sell rubbers and pencils . . . . Here are a few stationery stores I found . . . ” or “Erasers are called ‘rubbers’ in England. Here are a few stationery shops that sell rubbers and pencils.” By including the change in language or locale-specific vocabulary in the paraphrase, the digital assistant can provide some information to the user in context, without making the interaction with the user too cumbersome. Other similar examples include “gas” and “petrol,” “apartment” and “flat,” “can” and “tin,” “closet” and “wardrobe,” “elevator” and “lift,” etc.
In some embodiments, the first geographic area and the second geographic area are (624) both associated with a primary language of the user.
In some embodiments, the change in locale-specific vocabulary includes (626) use of a respective local slang in the second geographic area for the at least one word or expression in the second speech input, and wherein the second paraphrase utilizes the respective local slang.
In some embodiments, the digital assistant receives (628) user input to start a learning session regarding the respective local slang provided in the second paraphrase. In response to receiving the user input, the digital assistant provides (630) an explanation of the usage of the respective local slang in the second geographic area.
In some embodiments, the change in language includes (632) use of a respective local accent in the second geographic area for the at least one word or expression in the second speech input, and wherein the second paraphrase utilizes the respective local accent.
In some embodiments, the digital assistant receives (634) user input to start a learning session regarding the respective local accent provided in the second paraphrase. In response to receiving the user input, the digital assistant provides (636) one or more additional examples of the usage of the respective local accent in the second geographic area.
In some embodiments, the digital assistant further implements any combination of the features described above and in the remainder of this specification.
In the process 700, in some embodiments, the digital assistant evaluates (702) a present context associated with a user. The digital assistant identifies (704) a respective foreign language training scenario associate with the present context. The digital assistant then provides (706) a foreign language training session for the user, the foreign language training session containing one or more foreign language exercises tailored for the current context.
In an example scenario, when the user is in a foreign country, the digital assistant keeps track of the user's current location. When the user is located inside a grocery store in the foreign country, the digital assistant optionally generates foreign language exercises that are suitable for the current context. For example, the foreign language exercises may include vocabulary about food products (e.g., words for names of food products, categories of food products, description of food products, nutrition information of food products, prices, discounts, etc.). In addition, the foreign language exercises may also include dialogues and/or phrases related to interactions that commonly occur in a grocery store. For example, the digital assistant optionally generates foreign language exercises related to asking for help to locate a particular food item, asking for different alternatives for a food product, asking about origins or sources of a particular type of food product, asking about discounts for a particular product, and errors in prices for particular products and/or the total bill, etc.
In some embodiments, the digital assistant can carry out a foreign language dialogue with the user to practice various aspects of shopping in the grocery store. Dynamically generating foreign language exercises based on the current context is helpful for the user to learn a foreign language more quickly. The user may better remember words in the foreign language, when he or she can see in person the particular items that those words describe. Comprehension and memorization of the foreign language dialogues and vocabulary can also be enhanced when the user see the same words in context (e.g., food labels in the grocery store), and hear the dialogue spoken by native speakers in real life (e.g., other customers asking about prices and discounts in the grocery store).
In some embodiments, the digital assistant automatically, without user intervention, selects (708) a respective language for the one or more foreign language exercises based on a primary language associated with a geographic area in which the user is currently located. Then, the digital assistant generates (710) the one or more foreign language exercises in the automatically selected language.
In some embodiments, the digital assistant receives (712) user input selecting a respective language for the one or more foreign language exercises, and generates (714) the one or more foreign language exercises in the user-selected language.
In some embodiments, the present context associated with the user includes (716) the user's presence inside a store located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with shopping in the store.
In some embodiments, the present context associated with the user includes (718) the user's presence in proximity to a terminal of public transportation located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with use of the public transportation.
In some embodiments, the present context associated with the user includes (720) the user's presence inside a dining facility located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with dining at the dining facility.
In some embodiments, the present context associated with the user includes (722) the user's presence inside a lodging facility located in a geographic area in which a respective foreign language is a primary language, and the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with lodging at the lodging facility.
In some embodiments, the present context associated with the user includes (724) the user's presence inside a public transport vehicle moving toward a destination for which the user has recently requested directions and the destination is located in a geographic area in which a respective foreign language is a primary language, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with visiting to said destination.
In some embodiments, the present context associated with the user includes (726) the user's presence inside a healthcare facility, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with obtaining healthcare services at the healthcare facility.
In some embodiments, the present context associated with the user includes (728) the user's presence inside a business premise offering beverage services, and wherein the one or more foreign language exercises include at least vocabulary or dialogue in the respective foreign language that is associated with ordering beverages at the business premise.
In some embodiments, the digital assistant presents (730) images associated with vocabulary used in the foreign language exercises.
In some embodiments, the digital assistant further implements any combination of the features described above.
The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to best explain the principles of the invention and its practical applications, to thereby enable others skilled in the art to best utilize the invention and various embodiments with various modifications as are suited to the particular use contemplated.
This Application is a continuation of U.S. Nonprovisional application Ser. No. 14/213,852, filed on Mar. 14, 2014, entitled “USER TRAINING BY INTELLIGENT DIGITAL ASSISTANT”, which claims the benefit of U.S. Provisional Application No. 61/800,846, filed on Mar. 15, 2013, entitled “USER TRAINING BY INTELLIGENT DIGITAL ASSISTANT”, all of which are hereby incorporated by reference in their entity for all purposes.
| Number | Date | Country | |
|---|---|---|---|
| 61800846 | Mar 2013 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 14213852 | Mar 2014 | US |
| Child | 17473447 | US |
