1. Field of the Invention
The present invention relates to a feature for use in web-based services and in particular, to a system, method and computer-readable medium that provides automated web transcription services.
2. Introduction
Transcribing human speech is a multi billion dollar business in the United States alone. One of the major drivers for speech transcription is the medical industry. The majority of the medical transcription industry is outsourced to India for manual transcription and is worth over $40 billion dollar per year. Many companies provide medical professionals the ability to manually transcribe recorded files but charge about $4-5 per report (10 cents per 64 characters). The word accuracy of these reports range from 94-98% per word.
Conventional ways of performing automatic transcription of medical reports as well as other types of reports, such as legal and government documents, have included several attempts for creating stand-alone personal computer (PC) products for business professionals, such as the Nuance Dragon system. However, these programs have not been very successful primarily because (1) the failure of the speech technology in providing high accuracy speech recognition, (2) the complexity of using and customizing the speech technology for specific industrial or personal needs, and (3) the lack of flexibility provided by the fact that access to the dictation software and access to the dictation itself is limited to a single PC.
A system, method and computer readable medium that provides automated web transcription service is disclosed. The method may include receiving input speech from a user using a communications network, recognizing the received input speech, understanding the recognized speech, transcribing the understood speech to text, storing the transcribed text in a database, receiving a request via a web page to display the transcribed text, retrieving transcribed text from the database, and displaying the transcribed text to the requester using the web page.
In order to describe the manner in which the above-recited and other advantages and features of the invention can be obtained, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The features and advantages of the invention may be realized and obtained by means of the instruments and combinations particularly pointed out in the appended claims. These and other features of the present invention will become more fully apparent from the following description and appended claims, or may be learned by the practice of the invention as set forth herein.
Various embodiments of the invention are discussed in detail below. While specific implementations are discussed, it should be understood that this is done for illustration purposes only. A person skilled in the relevant art will recognize that other components and configurations may be used without parting from the spirit and scope of the invention.
The present invention comprises a variety of embodiments, such as a system, method, computer-readable medium, and other embodiments that relate to the basic concepts of the invention.
One of skill in the art will appreciate that the network environment 100 may be any possible configuration in which a processing device may be accessible to the internet 120. While the internet 120 is illustrated, one of skill in the art will appreciate that the web data language model generation process discussed herein may be implemented in conjunction with other network systems, such as an intranet or any other network system capable of hosting domains containing web page data, for example.
The network environment 100 and the automated web transcription service unit 130 illustrated in
The ASR 210 receives a user's input speech and may be of any known design and performs the function of recognizing, or spotting, the existence of one or more speech parts in the user's input speech. The ASR 210 may also be tasked with recognizing system commands, such as “CORRECTION”, “STOP”, “PAUSE”, “BEGIN”, “GO TO (parts of a customized template)”, etc.
The SLU unit 220 may be of any known design that performs the function of receiving and understanding the recognized speech and system commands that may be spoken by the user.
The transcription generator 230 receives the understanding data from the SLU unit 220 and forms the transcribed document for presentation to the user.
Processor 320 may include at least one conventional processor or microprocessor that interprets and executes instructions. Memory 330 may be a random access memory (RAM) or another type of dynamic storage device that stores information and instructions for execution by processor 320. Memory 330 may also store temporary variables or other intermediate information used during execution of instructions by processor 320. ROM 340 may include a conventional ROM device or another type of static storage device that stores static information and instructions for processor 320. Storage device 350 may include any type of media, such as, for example, magnetic or optical recording media and its corresponding drive.
Input device 360 may include one or more conventional mechanisms that permit a user to input information to the automated web transcription service unit 130, such as a keyboard, a mouse, a pen, a voice recognition device, etc. Output device 370 may include one or more conventional mechanisms that output information to the user, including a display, a printer, one or more speakers, or a medium, such as a memory, or a magnetic or optical disk and a corresponding disk drive. Communication interface 380 may include any transceiver-like mechanism that enables the automated web transcription service unit 130 to communicate via a network. For example, communication interface 380 may include a modem, or an Ethernet interface for communicating via a local area network (LAN). Alternatively, communication interface 380 may include other mechanisms for communicating with other devices and/or systems via wired, wireless or optical connections. In some implementations of the network environment 100, communication interface 380 may not be included in exemplary automated web transcription service unit 130 when the automated web transcription process is implemented completely within a single automated web transcription service unit 130.
The automated web transcription service unit 130 may perform such functions in response to processor 320 by executing sequences of instructions contained in a computer-readable medium, such as, for example, memory 330, a magnetic disk, or an optical disk. Such instructions may be read into memory 330 from another computer-readable medium, such as storage device 350, or from a separate device via communication interface 380.
For illustrative purposes, the web data language model generation process will be described below in relation to the block diagrams shown in
At step 4500, the transcription generator 230 transcribes the understood speech to text and generates a document in a suitable format for presentation and display. At step 4600, the transcription generator 230 send the transcribed document to the transcription database 240 for storage.
In step 4700, the transcription generator 230 receives a request via a web page to display the transcribed text. At step 4800, the transcription generator 230 retrieves the transcribed text from the web transcription database 240, and at step 4900, displays the transcribed text to the requester using the web page. The transcription may be displayed on the same web page or web site that was used to upload the transcription or request the transcription or it may be display on a different one. In any event, the requester may go to the website and review the transcribed dictation. There may be an approval process whereby the requester can approve or not approve the transcription. The requester may also be permitted to correct the transcription. Once the transcription is approved, a charge may be made on a credit card, for example. Alternatively, a subscription rate may be such that a yearly payment is made for unlimited transcriptions, for example.
The transcription generator 230 may allow the requester to simply print out the transcription. Alternatively, the transcription generator 230 may provide the option to e-mail, send etc. the transcription to another user or institution, for example. In this manner, one person can upload or dictate, another can approve, and still a third person can receive the finished transcription. Or it can be the same person performing all three processes.
As stated above, the transcription generator 230 may allow the requester correct errors in the transcription. The corrected errors may be fed back into the automated web transcription service unit 130 in an active learning manner to improve the future accuracy of the transcriptions. The process goes to step 4950 and ends.
The automated web transcription service unit 130 process may also allow users to enroll through the web. For example, a group of professionals (e.g., law firm or hospital) or a person such as a doctor on his/her own would go to the website and would enroll in the service by providing certain personal and billing information, such as credit card information, etc. The automated web transcription service unit 130 may also allow the user to identify from a simple drop down menu what the profession in which the individual or group specializes. For example, these categories may be broad, such as doctor, lawyer, realtor, etc., or more specific, such as intellectual property attorney, cardiologist, etc. In that regard, the accuracy of the transcription may increase if the ASR 210 and/or SLU unit 220 used is trained on the specific domain.
As stated above, the audio files (e.g., .wavs, etc.) for transcription may be uploaded using the web or through a toll-free, local, or long distance telephone number and a password, for example. The password can be a phrase, PIN or other number that is spoken or entered on a keypad, for example. Verification can also be performed by swiping a credit card or entering an account number.
The input provided by the user to automated web transcription service unit 130 may follow a template that is specific to their profession or practice. For example, in the medical profession, the system may provide standard structure for various medical reports or customized templates may be provided to the system in advance. So within the structure a doctor may decide that there are five sections for possible dictation. One section could be the patient's background, second section could be the patient's history, third one could be the patient's allergy to medications, fourth could be a recommendation for a diagnosis, the fifth section could be where to send the report. A doctor may choose to dictate all sections or just certain ones.
In that manner, a doctor may say “Patient's history” and begin dictating for that section or simply “GO TO section one”. After that section is complete, the doctor may give an “END OF SECTION” command or simply state the name of the next section on the template such as “diagnosis” and beginning dictating that section. Therefore, the user can jump to various sections in the template and skip sections altogether. Thus, as stated above, the automated web transcription service unit 130 may be able to recognize these sections as specific system commands and act on them apart from the dictation itself.
Since individual make mistakes, matter instructions or commands may be recognized so that the user can navigate through the dictation and correct mistakes. Commands such as “BACK”, “GO TO”, BEGINNING OF LINE”, etc. may be used and recognized by the ASR 210 and understood by the SLU unit 220.
The automated web transcription service unit 130 may require the user to train the ASR 210 according the individual's specific voice, language and/or dialect. The individual may spend time training the system or upload pre-recorded audio file which generate transcriptions can be review and corrected for feedback.
Therefore, the automated web transcription service unit 130 may use several models for an individual. For example, if the individual is a doctor, the automated web transcription service unit 130 may use an acoustic model, a language model, and a pronunciation model that are used to operate with his/her voice. The acoustic model extracts the features and quality of an individual's voice, as opposed from someone else's voice. The language model tracks the language or context used by the speaker and may customize the model to the individual's practice or profession. The pronunciation model tracks the personalized pronunciation features of words used by the individual.
In all of these models, the automated web transcription service unit 130 is trying to maximize the accuracy of the transcription not for a just any doctor, but for each individual doctor using the service. Therefore, the automated web transcription service unit 130 may track and adapt to the individual to provide a personalized model.
The automated web transcription service unit 130 may also provide the ability to go to a user interface for changing or correcting the transcription that may permit the user to click on any word or phrase and listen to what the dictation said for that particular word or phrase.
The automated web transcription service unit 130 may provide archival services for transcribed documents using the web transcription database 240, for example, so that an individual or intuition may go back several years and review note and files.
The automated web transcription service unit 130 may use a form of adaptive speech recognition. In particular, when a doctor dictates into the automated web transcription service unit 130 and then reviews the transcription, he or she may make corrections. However, the doctor may make corrections not because the automated web transcription service unit 130 made an error but because he or she wanted something reworded. So the end product is a final report that may not be an accurate transcription of a speech because items were changed. The adaptation occurs once the transcribed document is submitted where the automated web transcription service unit 130 identifies segments of what it feels confident that were linked to the changed words and it adapts. Thus, the automated web transcription service unit 130 is learning and adapting from all available information.
The automated web transcription service unit 130 may be configured to provide the user with a transcript that highlights words or phrases for which it did not attain a certain confidence level. The user could then review those words and provide feedback to the system to improve future accuracy.
The automated web transcription service unit 130 may also learn adaptive learning techniques from new data and human correction data that involve sharing the data from a larger user base. For example, the automated web transcription service unit 130 may utilize specific new and/or correction data from a group of people that are lawyers, a group that are doctors, etc. and make it available for the recognition and understanding benefit of all of the users in that particular category.
The automated web transcription service unit 130 may also track the accuracy of particular transcriptions. For example, if the transcription is found to be below a certain accuracy level, the user does not have to pay for the transcription. In this manner, the automated web transcription service unit 130 can track the number of corrections made to gauge the accuracy or record the ASR's 210 confidence level that words were correct.
Embodiments within the scope of the present invention may also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media that can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures. When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or combination thereof) to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such connection is properly termed a computer-readable medium. Combinations of the above should also be included within the scope of the computer-readable media.
Computer-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing device to perform a certain function or group of functions. Computer-executable instructions also include program modules that are executed by computers in stand-alone or network environments. Generally, program modules include routines, programs, objects, components, and data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps.
Although the above description may contain specific details, they should not be construed as limiting the claims in any way. Other configurations of the described embodiments of the invention are part of the scope of this invention. For example, the principles of the invention may be applied to each individual user where each user may individually deploy such a system. This enables each user to utilize the benefits of the invention even if some or all of the conferences the user is attending do not provide the functionality described herein. In other words, there may be multiple instances of the automated web transcription service unit 130 in
The present application is a continuation of U.S. patent application Ser. No. 13/975,901, filed Aug. 26, 2013, which is a continuation of U.S. patent application Ser. No. 11/469,016, filed Aug. 31, 2006, now U.S. Pat. No. 8,521,510, issued Aug. 27, 2013, the content of which is incorporated herein by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
5450524 | Rissanen | Sep 1995 | A |
5761689 | Rayson et al. | Jun 1998 | A |
5864805 | Chen et al. | Jan 1999 | A |
5907597 | Mark | May 1999 | A |
6104995 | Shimada et al. | Aug 2000 | A |
6122614 | Kahn et al. | Sep 2000 | A |
6173259 | Bijl et al. | Jan 2001 | B1 |
6347296 | Friedland | Feb 2002 | B1 |
6366882 | Bijl et al. | Apr 2002 | B1 |
6418410 | Nassiff et al. | Jul 2002 | B1 |
6789231 | Reynar et al. | Sep 2004 | B1 |
6865258 | Polcyn | Mar 2005 | B1 |
6901364 | Nguyen et al. | May 2005 | B2 |
6999933 | Hoi | Feb 2006 | B2 |
7069513 | Damiba | Jun 2006 | B2 |
7184956 | Frimpong-Ansah | Feb 2007 | B2 |
7292975 | Lovance et al. | Nov 2007 | B2 |
7321852 | Stanford | Jan 2008 | B2 |
7376561 | Rennillo et al. | May 2008 | B2 |
7502741 | Finke et al. | Mar 2009 | B2 |
7668718 | Kahn et al. | Feb 2010 | B2 |
7818175 | Carus et al. | Oct 2010 | B2 |
8032372 | Zimmerman et al. | Oct 2011 | B1 |
8117034 | Gschwendtner | Feb 2012 | B2 |
8229742 | Zimmerman et al. | Jul 2012 | B2 |
20020046032 | Wutte | Apr 2002 | A1 |
20020143533 | Lucas et al. | Oct 2002 | A1 |
20030014261 | Kageyama | Jan 2003 | A1 |
20030050778 | Nguyen et al. | Mar 2003 | A1 |
20030182622 | Sibal et al. | Sep 2003 | A1 |
20030225578 | Kahn et al. | Dec 2003 | A1 |
20040006474 | Gong et al. | Jan 2004 | A1 |
20040088162 | He et al. | May 2004 | A1 |
20040141597 | Giacomelli | Jul 2004 | A1 |
20040199375 | Ehsani et al. | Oct 2004 | A1 |
20040267527 | Creamer et al. | Dec 2004 | A1 |
20050010411 | Rigazio et al. | Jan 2005 | A1 |
20060041427 | Yegnanarayanan et al. | Feb 2006 | A1 |
20060041428 | Fritsch et al. | Feb 2006 | A1 |
20060074656 | Mathias et al. | Apr 2006 | A1 |
20060149558 | Kahn et al. | Jul 2006 | A1 |
20060190253 | Hakkani-Tur et al. | Aug 2006 | A1 |
20070106724 | Gorti et al. | May 2007 | A1 |
20070208567 | Amento et al. | Sep 2007 | A1 |
20070299664 | Peters et al. | Dec 2007 | A1 |
20090048833 | Fritsch et al. | Feb 2009 | A1 |
Number | Date | Country | |
---|---|---|---|
20140316780 A1 | Oct 2014 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 13975901 | Aug 2013 | US |
Child | 14321932 | US | |
Parent | 11469016 | Aug 2006 | US |
Child | 13975901 | US |