Patent Application
20190228789
The present technology relates generally to audio enhancement and, in particular, to voice or speech enhancement techniques to cancel noise heard/measured in a moving vehicle.
Road vehicles such as cars, trucks, vans, motorcycles, etc. are often equipped with various speech-based subsystems such as hands-free telephony and speech recognition. The perceptual quality of a speech signal can be significantly degraded by unwanted ambient noise from sources such as the powertrain, tires, wind and rain. Noise cancellation may be accomplished using microphones that detect noise. A speech enhancement processor can use the detected noise to suppress some of the noise to thereby enhance the speech signal. However, due to the lag in activating a speech-enhancement processor to provide noise cancellation, the speech-enhancement processor may not be able to react quickly enough to sudden changes in noise, thus degrading the quality of the speech signal. For example, a road with a rough surface, a grated bridge or a tunnel will all cause a sudden change in noise that may degrade the speech signal or other audio signal.
It would thus be highly desirable to mitigate the effect of sudden changes in noise in order to enhance a voice signal or other audio signal inside a moving vehicle. Improvements in speech and audio enhancement technology are thus highly desirable.
Further features and advantages of the present technology will become apparent from the following detailed description, taken in combination with the appended drawings, in which:
It will be noted that throughout the appended drawings like features are identified by like reference numerals.
Disclosed in this specification and illustrated in the drawings is an audio enhancement system, which may be more particularly a speech enhancement system (or voice enhancement system) for enhancing an audio signal such as a speech signal by suppressing unwanted noise that arises from the movement of a vehicle. In general, the system predicts an expected noise at a location and then prepares an audio enhancement processor (which may be a speech enhancement processor in one main implementation) to cancel or suppress noise based on the expected noise (or predicted noise) when passing by the location. By predicting the noise to be encountered, the system eliminates, or at least significantly mitigates, unwanted noise in an audio signal such as a speech signal. The enhanced audio signal can thus improve the quality of hands-free telephony, in-vehicle speech recognition, an in-vehicle sound system, or other application that utilizes the audio or speech signal.
Accordingly, one aspect of the present disclosure is an audio enhancement system for a vehicle that includes a location-determining subsystem configured to determine a first location of the vehicle and to generate location data representing the first location of the vehicle and an audio-enhancement processor for processing an audio signal. The processor is communicatively coupled to the location-determining subsystem for receiving the location data. The processor is configured to predict expected noise at a second location corresponding to a predicted location of the vehicle. The processor is further configured to suppress noise based on the expected noise when the vehicle is at the second location to thereby enhance the audio signal.
Another aspect of the present disclosure is a non-transitory computer-readable medium comprising instructions in code which when loaded into a memory and executed by an audio-enhancement processor of a vehicle cause the processor to receive location data representing a first location of the vehicle from a location-determining subsystem, predict expected noise at a second location corresponding to a predicted location of the vehicle, and suppress noise based on the expected noise when the vehicle is at the second location to thereby enhance an audio signal.
Yet another aspect of the present technology is a vehicle comprising an audio enhancement system that includes a location-determining subsystem configured to determine a first location of the vehicle and to generate location data representing the first location of the vehicle and an audio-enhancement processor for processing an audio signal. The processor is communicatively coupled to the location-determining subsystem for receiving the location data. The processor is configured to predict expected noise at a second location corresponding to a predicted location of the vehicle. The processor is further configured to suppress noise based on the expected noise when the vehicle is at the second location to thereby enhance the audio signal.
The details and particulars of these aspects of the technology will now be described below, by way of example, with reference to the drawings.
As illustrated by way of example in
In the embodiment depicted by way of example in
In the embodiment depicted in
In the embodiment illustrated in
In one embodiment, the audio/speech-enhancement system 110 cancels noise based solely on the predicted or expected noise (without measuring any actual in-cabin noise using the microphones 120). In another embodiment, the speech-enhancement system 110 cancels noise based on both the actual in-cabin noise measured by the microphones 120 and the expected/predicted noise.
In the embodiment depicted in
In one embodiment, the memory 114 coupled to the processor 112 is used to store noise data for a route. The noise data may be collected by the vehicle or obtained from an external source, e.g. another vehicle, as will be described below. Noise data may be stored for a plurality of locations along a route. Noise data may include a noise frequency spectrum and location coordinates. In a variant, the noise data may be time-specific as well. For example, some noise may only occur at certain times of day, e.g. a construction site that is only operational during the daytime but quiet at night. Noise data may be stored for frequently traveled routes such as a route between work and home. In one particular embodiment, the processor is configured to identify noise patterns in the route to thus learn the noise profile along a given route to thus be able to predict the noise to be expected at various points along the route. In other words, the processor may be configured to predict the expected noise based on the noise patterns in the route. As such, the processor is able to learn what to expect in terms of noise along a frequently traveled route. For example, if the route from home to work takes the vehicle through a tunnel, over a grated bridge, and past a construction site with loud jackhammers, the processor records these noises and notes their locations. On a subsequent drives to work, the processor predicts the noise to be encountered in the tunnel, over the grated bridge and past the construction site. As the vehicle approaches each of the predictably noisy locations, the audio/speech-enhancement system 110 predicts the noise that is imminently expected and enhances the audio/speech signal by suppressing the noise just as the vehicle passes the noisy location.
In one embodiment, the vehicle further includes a radio frequency transceiver 150, e.g. a cellular data transceiver, which is coupled to the processor for downloading noise patterns from a shared server 160, which may also be a server cluster, server farm or any cloud-based data storage system. The shared server 160 may be accessible via the internet 200 through a cellular base station transceiver 210 at a nearby cell tower or through other access points (not shown). Noise data may be uploaded and downloaded, i.e. shared, amongst other users. Noise data may be grouped according to vehicle type. Noise data may be downloaded for new routes yet to be traveled. The download may occur automatically or in response to a user command. For example, upon programming a destination in a navigation system, the noise data for the route may be downloaded.
In some embodiments, the enhancement is performed on a speech signal obtained from one or more the microphones within the vehicle (e.g. a speech signal derived from a telephone call or from voice commands spoken in the context of hands-free control of vehicle functions). In other embodiments, the enhancement techniques described above may be applied to an audio signal carrying audio content played by the speakers into the car in order to counteract the unwanted noise. The audio signal (of the audio content) may come from, for example, the radio, compact disc (CD) player, MP3 player, onboard hard disk containing digital audio media, or music streamed from a vehicle-connected mobile device. By predicting the amount of noise at an upcoming location, the audio signal can be enhanced using the predicted or expected amount of noise at the upcoming location. For the purposes of this specification, the expression “audio” is meant to encompass “speech”.
In at least one embodiment, the processor of the system is coupled to a memory to store noise data for a route. In this or other embodiments, the processor is configured to identify noise patterns in the route. In any of the foregoing embodiments, the processor is configured to predict the expected noise based on the noise patterns in the route. In any of the foregoing embodiments, the system further has a radiofrequency transceiver coupled to the processor for downloading noise patterns from a shared server. In any of the foregoing embodiments, the location-determining subsystem is a Global Navigation Satellite System receiver.
The computer-readable medium described above may further comprise code to cause the memory to store noise data for a route. In this or other embodiments, the computer-readable medium may further have code to cause the processor to identify noise patterns in the route. In any of the foregoing embodiments, the computer-readable medium further includes code to predict the expected noise based on the noise patterns in the route. In any of the foregoing embodiments, the computer-readable medium further includes code to instruct a radiofrequency transceiver to download noise patterns from a shared server.
In the vehicle described, the processor may be coupled to a memory to store noise data for a route. In this or other embodiments, the processor may be configured to identify noise patterns in the route. In any of the foregoing embodiments, the processor is optionally configured to predict the expected noise based on the noise patterns in the route. In any of the foregoing embodiments, the vehicle includes a radiofrequency transceiver coupled to the processor for downloading noise patterns from a shared server. In any of the foregoing embodiments, the location-determining subsystem of the vehicle is a Global Navigation Satellite System receiver.
Any of the methods disclosed herein may be implemented in hardware, software, firmware or any combination thereof. Where implemented as software, the method steps, acts or operations may be programmed or coded as computer-readable instructions and recorded electronically, magnetically or optically on a fixed, permanent, non-volatile or non-transitory computer-readable medium, computer-readable memory, machine-readable memory or computer program product. In other words, the computer-readable memory or computer-readable medium comprises instructions in code which when loaded into a memory and executed on a processor of a computing device cause the computing device to perform one or more of the foregoing method(s).
A computer-readable medium can be any means that contain, store, communicate, propagate or transport the program for use by or in connection with the instruction execution system, apparatus or device. The computer-readable medium may be electronic, magnetic, optical, electromagnetic, infrared or any semiconductor system or device. For example, computer executable code to perform the methods disclosed herein may be tangibly recorded on a computer-readable medium including, but not limited to, a floppy-disk, a CD-ROM, a DVD, RAM, ROM, EPROM, Flash Memory or any suitable memory card, etc. The method may also be implemented in hardware. A hardware implementation might employ discrete logic circuits having logic gates for implementing logic functions on data signals, an application-specific integrated circuit (ASIC) having appropriate combinational logic gates, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
It is to be understood that the singular forms “a,” “an,” and “the” include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to “a processor” includes reference to one or more of such processors.
This invention has been described in terms of specific embodiments, implementations and configurations which are intended to be exemplary only. Persons of ordinary skill in the art will appreciate, having read this disclosure, that many obvious variations, modifications and refinements may be made without departing from the inventive concept(s) presented herein. The scope of the exclusive right sought by the Applicant(s) is therefore intended to be limited solely by the appended claims.
