Abstract: An apparatus for reducing belt slip of a vehicle, in which the vehicle includes an engine as a vibration source and an electric motor connected to the engine through a belt to transmit torque of the engine, includes: a signal generator configured to generate a reference signal with a frequency corresponding to vibration of the engine; an adaptive filter configured to calculate a filter coefficient to remove an error value between a rotational speed of the engine and a rotational speed of the electric motor and apply the filter coefficient to the reference signal to generate a reference torque signal; and a torque compensator configured to generate a belt slip compensation torque signal by changing an amplitude of the reference torque signal, and apply the belt slip compensation torque signal to determine a final torque command of the electric motor.

Abstract: An apparatus for decoding an audio signal is provided, having a receiving interface, configured to receive a first frame having a first audio signal portion of the audio signal, and configured to receive a second frame having a second audio signal portion of the audio signal; a noise level tracing unit, wherein the noise level tracing unit is configured to determine noise level information depending on at least one of the first audio signal portion and the second audio signal portion; a first reconstruction unit for reconstructing, in a first reconstruction domain, a third audio signal portion of the audio signal depending on the noise level information; a transform unit for transforming the noise level information to a second reconstruction domain; and a second reconstruction unit for reconstructing, in the second reconstruction domain, a fourth audio signal portion of the audio signal depending on the noise level information.

Abstract: An apparatus for determining a reliability of a recommendation based on an environment of a vehicle may include: an input device configured to detect a sound in the vehicle; a storage configured to store information associated with the sound; an output device configured to output information associated with a recommendation action; and a controller electrically connected with the input device, the storage, and the output device.

Abstract: A system for exploiting visual information for enhancing audio signals via source separation and beamforming is disclosed. The system may obtain visual content associated with an environment of a user, and may extract, from the visual content, metadata associated with the environment. The system may determine a location of the user based on the extracted metadata. Additionally, the system may load, based on the location, an audio profile corresponding to the location of the user. The system may also load a user profile of the user that includes audio data associated with the user. Furthermore, the system may cancel, based on the audio profile and user profile, noise from the environment of the user. Moreover, the system may include adjusting, based on the audio profile and user profile, an audio signal generated by the user so as to enhance the audio signal during a communications session of the user.

Abstract: In one example of the disclosure, microphone data indicative of a user spoken phrase is captured utilizing a microphone. The microphone data is analyzed to detect a trigger phrase. A wakeup event is performed responsive to the detection of the trigger phrase. At least a portion of the microphone data is sent to the computing device for the computing device to send to a server system.

Abstract: The present invention is a method and apparatus for a first user to purchase a controlled product. The first user creates an order and specifies the set of conditions that the second user must meet to satisfy the order. The second user satisfies those conditions then can access an authorization from a third user prior to submitting a bid to supply the order. The first user can accept the bid then authorize payment. The first user and second user can confirm the product was delivered to the first user.

Abstract: Systems and methods that enable implementation of content aware file management labeling. Techniques disclosed enable real-time analysis of a data file so that associations between the data file and applicable file management label(s) can be automatically suggested and/or made while a user is working in the data file. A user may deploy an application on a client device to edit a data file. While the user is actively editing the data file, the application may transmit modification data to a content analysis service which analyzes the modification data to determine whether the modifications result in a file management label becoming applicable to the data file. Ultimately, the content analysis service may transmit a verdict to the client device to cause the application to display a labeling suggestion to the user and/or to automatically apply a label to the data file while it is being worked on by the user.

Abstract: Apparatuses and methods for selecting, for communicating, between signals provided by various pickups on the basis of a calculated signal to noise ratio (SNR) are disclosed. The various pickups may include a microphone and a vibration sensor. Signals from the microphone may be compared with signals from the vibration sensor by extracting a root-mean-square (RMS) profile for each, and comparing the RMS profiles to determine an SNR estimate for the microphone signal. The microphone signal may be selected if the SNR estimate is below a predetermined threshold, otherwise the vibration sensor signal may be selected. The vibration sensor signal may be subject to further processing if selected to approximate the microphone signal.

Abstract: A system and method for efficient universal background model (UBM) training for speaker recognition, including: receiving an audio input, divisible into a plurality of audio frames, wherein at least a first audio frame of the plurality of audio frames includes an audio sample having a length above a first threshold extracting at least one identifying feature from the first audio frame and generating a feature vector based on the at least one identifying feature; generating an optimized training sequence computation based on the feature vector and a Gaussian Mixture Model (GMM), wherein the GMM is associated with a plurality of components, wherein each of the plurality of components is defined by a covariance matrix, a mean vector, and a weight vector; and updating any of the associated components of the GMM based on the generated optimized training sequence computation.

Abstract: One embodiment provides a method, including: receiving, at a first device, a user input; activating, from the first device, a second device; and thereafter coordinating, using the processor of the first device, with the second device to process the user input. Other aspects are described and claimed.

Abstract: An embodiment of the present invention provides an apparatus for noise canceling that includes: an input unit configured to receive an input voice signal; and one or more processors configured to perform a first noise cancellation using as input the received input voice signal to generate a first voice signal by cancelling noise from the input voice signal using a noise canceling model which is trained using a plurality of reference voice signals, perform a second noise cancellation using as input the first voice signal generated by the noise canceling model to generate a second voice signal in which residual noise is canceled from the first voice signal using statistical analysis, and generate an output voice signal comprising an encoding of the second voice signal.

Abstract: The present disclosure relates to a method and system for controlling loudness of an audio based on signal analysis and deep learning. The method includes analyzing an audio characteristic in a frame level based on signal analysis, analyzing the audio characteristic in the frame level based on learning, and controlling loudness of the audio in the frame level, by combining the analysis results. Accordingly, reliability of audio characteristic analysis can be enhanced and audio loudness can be optimally controlled.

Abstract: A method for improving an echo cancellation effect and a system thereof are disclosed. The method comprises includes: performing a non-linear compensation for a non-linear response portion of an excitation signal using an NLC algorithm to obtain a compensated excitation signal; outputting the compensated excitation signal to an echo cancellation system; and performing echo cancellation for the compensated excitation signal by the echo cancellation system. According to the present disclosure, using the NLC algorithm, non-linear compensation is performed for the non-linear portion of the excitation signal, non-linear outputs generated due to non-linear characteristics of the system are pre-compensated when being input to the echo cancellation system, such that the echo signal output by the echo cancellation system is minimized and the echo cancellation effect is improved.

Abstract: In some embodiments, a pitch filter for filtering a preliminary audio signal generated from an audio bitstream is disclosed. The pitch filter has an operating mode selected from one of either: (i) an active mode where the preliminary audio signal is filtered using filtering information to obtain a filtered audio signal, and (ii) an inactive mode where the pitch filter is disabled. The preliminary audio signal is generated in an audio encoder or audio decoder having a coding mode selected from at least two distinct coding modes, and the pitch filter is capable of being selectively operated in either the active mode or the inactive mode while operating in the coding mode based on control information.

Abstract: A system may perform speech enhancement of audio data in real-time by suppressing noise components that are present in the audio data while preserving speech components. The system may include an in-ear module and a separate signal processing module that is wirelessly communicatively coupled to the in-ear module. The system may include non-negative matrix factorization (NMF) dictionaries capable of identifying frequency band components associated with speech and frequency band components associated with noise. The NMF dictionaries may be trained using voice samples and noise samples. The NMF dictionaries may be applied to noisy speech data to produce an NMF representation of the speech data which may then be applied using a dynamic mask to the noisy speech data in order to suppress the noise components of the noisy speech data and produce speech enhanced data.

Abstract: Embodiments of the invention are directed to apparatuses, methods, and computer program products for generating a rating score based on color. In some embodiments, the apparatuses, methods, and computer program products are configured to: receive at least one input, wherein the at least one input comprises at least one of a first numerical value, an emoticon, a color, or at least one keyword; process the at least one input; and generate a second numerical value based on processing the at least one input.

Abstract: Described embodiments include apparatus that includes a network interface (28) and a processor (30). The processor is configured to receive, via the network interface, speech of a subject (22) who suffers from a pulmonary condition related to accumulation of excess fluid, to identify, by analyzing the speech, one or more speech-related parameters of the speech, to assess, in response to the speech-related parameters, a status of the pulmonary condition, and to generate, in response thereto, an output indicative of the status of the pulmonary condition. Other embodiments are also described.

Abstract: A computer-implemented method, computer program product, and computer processing system are provided for word embedding. The method includes receiving, by a processor device, a word embedding matrix. The method further includes generating, by a processor device, an average pooling vector and a max pooling vector, based on the word embedding matrix. The method also includes generating, by the processor device, a prediction by applying a Multi-Layer Perceptron (MLP) to the average pooling vector and the max pooling vector.

Abstract: Audio information defining audio content may be accessed. The audio content may have a duration. The audio content may be segmented into audio segments. Individual audio segments may correspond to a portion of the duration. Feature vectors of the audio segments may be determined. The feature vectors may be processed through a classifier. The classifier may output scores on whether the audio segments contain voice. One or more of the audio segments may be identified as containing voice based on the scores and a two-step hysteresis thresholding. Storage of the identification of the one or more of the audio segments as containing voice in one or more storage media may be effectuated.

Abstract: Systems and methods are introduced to perform noise suppression of an audio signal. The audio signal includes foreground speech components and background noise. The foreground speech components correspond to speech from a user's speaking into an audio receiving device. The background noise includes babble noise that includes speech from one or more interfering speakers. A soft speech detector determines, dynamically, a speech detection result indicating a likelihood of a presence of the foreground speech components in the audio signal. The speech detection result is employed to control, dynamically, an amount of attenuation of the noise suppression to reduce the babble noise in the audio signal. Further processing achieves a more stationary background and reduction of musical tones in the audio signal.

Abstract: A method of audio source separation includes steps of applying a demixing matrix on a plurality of received signals to generate a plurality of separated results; performing a recognition operation on the plurality of separated results to generate a plurality of recognition scores; generating a constraint according to the plurality of recognition scores; and adjusting the demixing matrix according to the constraint; where the adjusted demixing matrix is applied to the plurality of received signals to generate a plurality of updated separated results from the plurality of received signals.

Abstract: A driving support method for a vehicle includes: acquiring a sound of a sound source placed outside the vehicle; and displaying, on a display portion, driving support information corresponding to an utterance content of the sound, the utterance content of the sound being recognized by a sound recognition process, in a display mode suggesting a relative position of the sound source from the vehicle, the relative position being specified based on the sound.

Abstract: A method for use in speaker verification, comprises: receiving a sample of a person's speech; estimating a noise measure in the received sample of the person's speech; if the estimated noise measure does not meet a predetermined criterion, rejecting the received sample of the person's speech; and if the estimated noise measure meets the predetermined criterion: adding noise to the received sample to form a sample with increased noise; extracting features of the sample with increased noise; and forming a model of the person's speech from the extracted features.

Abstract: A method includes obtaining closed-loop data associated with operation of an industrial process controller, where the industrial process controller is configured to control at least part of an industrial process using at least one model. The method also includes generating at least one noise model associated with the industrial process controller using at least some of the closed-loop data. The method further includes filtering the closed-loop data based on the at least one noise model. In addition, the method includes generating one or more model parameters for the industrial process controller using the filtered closed-loop data.

Abstract: A system configured to improve spatial coverage of output audio and a corresponding user experience by performing upmixing and loudspeaker beamforming to stereo input signals. The system can perform upmixing to the stereo (e.g., two channel) input signal to extract a center channel and generate three-channel audio data. The system may then perform loudspeaker beamforming to the three-channel audio data to enable two loudspeakers to generate output audio having three distinct beams. The user may interpret the three distinct beams as originating from three separate locations, resulting in the user perceiving a wide virtual sound stage despite the loudspeakers being spaced close together on the device.

Abstract: Various examples are provided related to speech recognition. In one example, a method includes converting an auditory signal into a pulse train, segmenting the pulse train into a series of frames having a predefined duration, and identifying a portion of the auditory signal by applying at least a portion of the series of frames segmented from the pulse train to a kernel adaptive autoregressive-moving-average (KAARMA) network. In another example, a speech recognition system includes processing circuitry configured to convert an auditory signal into a pulse train, segment the pulse train into a secured of frames, and identifying a portion of the auditory signal by applying at least a portion of the series of frames segmented from the pulse train to a KAARMA network. The series of frames segmented from the pulse train can be applied to a KAARMA chain including a plurality of KAARMA networks for identification.

Abstract: Active noise control (ANC), including active and adaptive noise cancellation (ANC) with non-voice-coil transducers having highly linear transfer functions, such as planar transducers, planar magnetic transducers, electro-static transducers, and piezo-electric transducers. This active and adaptive noise cancellation (ANC) may be used with: planar transducer headphones and earphones; open-backed and closed-back headphones and earphones; in-ear earphones, and phase plugs.

Abstract: An apparatus for reducing cross-talk between transmitted audio signals and received audio in a headset. The headset includes one or more of a set of earphones, a headset frame, a microphone boom with an array of MEMS microphone configured to isolate the earphone audio from the microphone audio, a VOX circuit, low crosstalk cable(s), and/or other components. Sets of microphones may be enabled and/or disabled to reduce cross-talk between received audio signals and transmitted audio signals. The VOX circuit is configured to reduce cross-talk between received audio signals and transmitted audio signals.

Abstract: Examples disclosed herein provide the ability for a computing device to determine a noise threshold to wake on ambient noises. In one example method, the computing device tracks sound, detected by a microphone of the computing device, over a period of time and, based on the sound tracked over the period of time, determines a noise threshold. The computing device tunes a sensitivity of the microphone to wake the computing device when ambient noise, detected by the microphone, is to have a signal strength equal to or exceeding the noise threshold.

Abstract: There is provided a signal processing apparatus including: a control unit that executes, on a basis of a waveform signal generated in accordance with a motion of an attachment portion of a sensor attached to a tool or a body, effect processing for the waveform signal or another waveform signal, the waveform signal being output from the sensor. The signal processing apparatus performs presentation so that a body motion itself can be aurally felt.

Abstract: A wireless earpiece includes a wireless earpiece housing, at least one microphone for detecting ambient environment sound, and a processor disposed within the wireless earpiece housing, the processor configured to distinguish between two or more sources of sound within the ambient environment sound. The wireless earpiece further includes a user interface operatively connected to the processor. The processor is configured to receive user input through the user interface to select one of the sources of sound within the ambient environment sound and wherein the processor is configured to process the ambient environment sound to emphasize portions of the ambient environment sound generated by the one of the sources of the ambient environment sound selected by the user to produce a modified sound. The earpiece may further include a speaker operatively connected to the processor to reproduce the modified sound.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for evaluating the quality of a communication session. One of the methods includes identifying, by a communication system, a communication session between one or more users of the communication system, wherein, during the communication session, session data is routed between a first communications device of a first user of the communication system and one or more other communications devices along a communication path; obtaining, from each of a plurality of communication nodes along the communication path, quality data relating to a quality of the communication session at the communication node; generating, using the quality data, a model input to a quality score machine learning model; and providing the model input as input to the quality score machine learning model to generate the estimated quality score for at least the portion of the communication session.

Abstract: A dialog content is generated using information that is unique to a user and information that is not unique. The processing executed by a dialog system includes a step of identifying a person based on a dialog with a user, a step of acquiring personal information, a step of analyzing the dialog, a step of extracting an event, a step of searching for a local episode and a global episode based on the personal information and the event, a step of generating dialog data using the search result, a step of outputting a dialog, and a step of accepting user evaluation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for designating certain voice commands as hotwords. The methods, systems, and apparatus include actions of receiving a hotword followed by a voice command. Additional actions include determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, where a voice command that is designated as a hotword is treated as a voice input regardless of whether the voice command is preceded by another hotword. Further actions include, in response to determining that the voice command satisfies one or more predetermined criteria associated with designating the voice command as a hotword, designating the voice command as a hotword.

Abstract: Performing speech recognition in a multi-device system includes receiving a first audio signal that is generated by a first microphone in response to a verbal utterance, and a second audio signal that is generated by a second microphone in response to the verbal utterance; dividing the first audio signal into a first sequence of temporal segments; dividing the second audio signal into a second sequence of temporal segments; comparing a sound energy level associated with a first temporal segment of the first sequence to a sound energy level associated with a first temporal segment of the second sequence; based on the comparing, selecting, as a first temporal segment of a speech recognition audio signal, one of the first temporal segment of the first sequence and the first temporal segment of the second sequence; and performing speech recognition on the speech recognition audio signal.

Abstract: A system and method of updating automatic speech recognition parameters on a mobile device are disclosed. The method comprises storing user account-specific adaptation data associated with ASR on a computing device associated with a wireless network, generating new ASR adaptation parameters based on transmitted information from the mobile device when a communication channel between the computing device and the mobile device becomes available and transmitting the new ASR adaptation data to the mobile device when a communication channel between the computing device and the mobile device becomes available. The new ASR adaptation data on the mobile device more accurately recognizes user utterances.

Abstract: A speech signal recognition method, apparatus, and system. The speech signal recognition method may include obtaining by or from a terminal an output of a personalization layer, with respect to a speech signal provided by a user of the terminal, having been implemented by input of the speech signal to the personalization layer, the personalization layer being previously trained based on speech features of the user, implementing a global model by providing the obtained output of the personalization layer to the global model, the global model being configured to output a phonemic signal indicating a phoneme included in the speech signal through the global model being previously trained based on speech features common to a plurality of users, and re-training the personalization layer based on the phonemic signal output from the global model, where the personalization layer and the global model collectively represent an acoustic model.

Abstract: A second device that is in communication with a first device receives transmissions of the first device and detects a SILENCE period status of the first device, which corresponds to a status wherein the first device has no speech samples to be transmitted towards the second apparatus. The second device determines the type of the received transmissions, counts the number of received transmissions of a first type, and times a time interval between the last received transmission of the first type and the last received transmission of a second determined type. At reception of a transmission of the first type, the second device detects whether the first device is in the SILENCE period status on an evaluation of the counted number of transmissions of the first type and the time interval of the last received transmission of the first type and the last received transmission of the second type.

Abstract: A road noise cancellation (RNC) system may include a controller and attenuator for reducing the audibility of the noise floor caused by the system's vibration sensors. A level of anti-noise at a location in a passenger cabin that may be attributed to the sensor noise floor may be estimated. An actual sound level in the passenger cabin may be measured or estimated, with the sensor noise floor component algorithmically removed. The difference in levels may be compared to a predetermined threshold to determine an amount of attenuation, if any, to be applied to an anti-noise signal to reduce audibility.

Abstract: According to one aspect, a method for determining voice activity is disclosed, the method including receiving a frame of an input audio signal, the input audio signal having a sample rate, and spitting the audio signal into a plurality of subbands, the plurality of subbands including at least a lowest subband and a highest subband. The method further comprises filtering the lowest subband to reduce an energy of the lowest subband, estimating a noise level for at least some of the plurality of subbands, and computing a signal-to-noise ratio for at least some of the plurality of subbands. The method also includes determining a speech activity level based at least in part on the computed signal-to-noise ratios and an average of an energy of at least some of the plurality of subbands.

Abstract: Systems and methods are disclosed. A digitized human vocal expression of a user and digital images are received over a network from a remote device. The digitized human vocal expression is processed to determine characteristics of the human vocal expression, including: pitch, volume, rapidity, a magnitude spectrum identify, and/or pauses in speech. Digital images are received and processed to detect characteristics of the user face, including detecting if one or more of the following is present: a sagging lip, a crooked smile, uneven eyebrows, and/or facial droop. Based at least on part on the human vocal expression characteristics and face characteristics, a determination is made as to what action is to be taken. A cepstrum pitch may be determined using an inverse Fourier transform of a logarithm of a spectrum of a human vocal expression signal. The volume may be determined using peak heights in a power spectrum of the human vocal expression.

Abstract: The present disclosure relates to speech recognition systems and methods using an adaptive incremental learning approach. More specifically, the present disclosure relates to adaptive incremental learning in a self-taught vocal user interface.

Abstract: A speech recognition system utilizing automatic speech recognition techniques such as end-pointing techniques in conjunction with beamforming and/or signal processing to isolate speech from one or more speaking users from multiple received audio signals and to detect the beginning and/or end of the speech based at least in part on the isolation. Audio capture devices such as microphones may be arranged in a beamforming array to receive the multiple audio signals. Multiple audio sources including speech may be identified in different beams and processed.

Abstract: Many headsets include automatic noise cancellation (ANC) which dramatically reduces perceived background noise and improves user listening experience. Unfortunately, the voice microphones in these devices often capture ambient noise that the headsets output during phone calls or other communication sessions to other users. In response, many headsets and communication devices provide manual muting circuitry, but users frequently forget to turn the muting on and/or off, creating further problems as they communicate. To address this, the present inventors devised, among other things, an exemplary headset that detects the absence or presence of user speech, automatically muting and unmuting the voice microphone without user intervention. Some embodiments leverage relationships between feedback and feedforward signals in ANC circuitry to detect user speech, avoiding the addition of extra hardware to the headset.

Abstract: A method includes receiving, by sensors inside an enclosure of a vehicle, signals generated by signal generators in the enclosure of the vehicle. One of the sensors or signal generators may be part of a mobile device inside the enclosure. The method also includes determining a location and orientation of the mobile device from the signals. The method further includes determining, based on the location and orientation of the mobile device, an object in the enclosure that the mobile device is pointing to. The mobile device further includes transmitting a message to the mobile device in response to determining that the mobile device is pointing to the object, so as to cause the mobile device to display a user interface to allow the mobile device to control the object.

Abstract: A purpose of the present invention is to provide a technique for easily performing accurate voice recognition.

Abstract: Audio information defining audio content may be accessed. The audio content may have a duration. The audio content may be segmented into audio segments. Individual audio segments may correspond to a portion of the duration. Feature vectors of the audio segments may be determined. The feature vectors may be processed through a classifier. The classifier may output scores on whether the audio segments contain voice. One or more of the audio segments may be identified as containing voice based on the scores and a two-step hysteresis thresholding. Storage of the identification of the one or more of the audio segments as containing voice in one or more storage media may be effectuated.

Abstract: An apparatus for generating an enhanced signal from an input signal, wherein the enhanced signal has spectral values for an enhancement spectral region, the spectral values for the enhancement spectral regions not being contained in the input signal, includes a mapper for mapping a source spectral region of the input signal to a target region in the enhancement spectral region, the source spectral region including a noise-filling region; and a noise filler configured for generating first noise values for the noise-filling region in the source spectral region of the input signal and for generating second noise values for a noise region in the target region, wherein the second noise values are decorrelated from the first noise values or for generating second noise values for a noise region in the target region, wherein the second noise values are decorrelated from first noise values in the source region.

Abstract: A method for reducing noise to a user to enable a conversation-of-interest to be heard, the noise originating from a noise source, the method comprising the steps of: operating at least one first device located at a first distance from the noise source, the user having noise-cancellation earphones connected to a second mobile device, the second mobile device located at a second distance from the noise source, the first distance less than the second distance; prehearing noise from the noise source using the at least one first device; analyzing the preheard noise to yield a respective analyzed noise signal; and processing the respective analyzed noise signal to effect noise cancellation for the noise-cancellation earphones.

Abstract: Aspects of the subject matter described in this disclosure can be implemented in a fall detection device and method. One or more motion sensors can access a user's acceleration data. The acceleration data can be segmented using a segmentation algorithm to identify a potential fall event. The segmentation algorithm can determine a cumulative sum of the acceleration data, where the cumulative sum is based on acceleration values being greater than or less than an acceleration threshold value, and a potential fall event can be identified where the cumulative sum is greater than a cumulative sum threshold value. Statistical features can be extracted from the segmented acceleration data and aggregated, and a determination can be made as to whether the potential fall event is a fall event based at least in part on the statistical features.