Abstract: A hearing device includes: an input module for provision of a first input signal; a processor configured to provide an electrical output signal based on the first input signal; a receiver configured to provide an audio output signal; and a controller comprising a speech intelligibility estimator configured to determine a speech intelligibility indicator indicative of speech intelligibility based on the first input signal, wherein the controller is configured to control the processor based on the speech intelligibility indicator; wherein the speech intelligibility estimator comprises a decomposition module configured to decompose the first input signal into a first representation of the first input signal in a frequency domain, wherein the first representation comprises one or more elements representative of the first input signal; and wherein the decomposition module comprises one or more characterization blocks for characterizing the one or more elements of the first representation in the frequency domain.

Abstract: A method for encoding multiple descriptions for a media stream includes: determining, for a current block of the media stream, a first number of duplicate blocks, determining, for the current block, a plurality of complementary forward gain factors, and determining a first number of descriptions for the current block by applying the plurality of complementary forward gain factors to the first number of duplicate blocks. A method for decoding multiple descriptions for a media stream includes: determining, for a current block of the media stream, a first number of descriptions from the media stream, in which each description comprises a scaled block corresponding to the current block, determining backward gain factors for the first number of scaled blocks, and determining the current block by applying the backward gain factors to the first number of scaled blocks.

Abstract: Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) may receive, from a base station in a full duplex mode, a first downlink transmission associated with a spatial filter. The UE may decode the first downlink transmission according to a default transmission configuration indicator (TCI) state that indicates the spatial filter. The default TCI state may be based at least in part on a first active TCI state for the full duplex mode. Numerous other aspects are provided.

Abstract: An audio data processing method, an electronic device, and a storage medium are provided, which are related to the field of data processing. The method includes: receiving sound production requests respectively sent by a plurality of sound production applications, wherein respective sound production requests include information of ports corresponding to respective sound production applications; acquiring audio data and sound mixing parameters corresponding to the respective sound production applications through the ports corresponding to the respective sound production applications; and mixing respective audio data according to the sound mixing parameters for the respective sound production applications, to generate audio data to be played.

Abstract: Disclosed is a wireless earphone linking method that includes the following: detecting a motion state of a wireless earphone; when the wireless earphone is moved, determining whether a working mode of the wireless earphone is a wearing mode; if so, determining whether the wireless earphone and a terminal are in a linked state; and if the wireless earphone and the terminal are not in a linked state, controlling the wireless earphone to send a linking request to the terminal.

Abstract: A beam generator, a beam generating method, and a chip are provided. The beam generator comprises a first channel, a second channel, and a signal merging module; the first channel comprises a first-channel filter, the first-channel filter is used to filter an input signal to obtain a first filtered signal; the first filtered signal comprises a desired signal; the second channel comprises: a second-channel blocking module, used to block the desired signal in the input signal to obtain a blocked signal; a compensation filter, connected to the second-channel blocking module for compensating for the blocked signal to obtain a second filtered signal; and an adaptive filter connected to the compensation filter for adaptively filtering the second filtered signal to obtain a third filtered signal; the signal merging module is for merging the first filtered signal and the third filtered signal to obtain an output signal.

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: A receiver included in a memory device includes a flag generator circuit, an equalizer circuit and an equalization controller circuit. The flag generator circuit is configured to, during a normal operation mode, generates a flag signal without an external command. The equalizer circuit is configured to, during the normal operation mode, receive an input data signal through a channel, generate an equalized signal by equalizing the input data signal based on an equalization coefficient, and generate a data sample signal including a plurality of data bits based on the equalized signal. The equalization controller circuit is configured to, during the normal operation mode, determine an amount of change in the equalization coefficient based on the flag signal, the equalized signal and the data sample signal, and perform a training operation in which the equalization coefficient is updated in real time based on the amount of change in the equalization coefficient.

Abstract: A method comprising determining feature values of an input audio window and determining a formant attenuation/amplification coefficient for the input audio window based on the processing of the feature values by a neural network.

Abstract: A method for hybrid noise suppression. The method involves obtaining a noisy audio input signal, generating a noise-suppressed audio output signal by performing a combination of a noise spectrum estimate-based noise suppression and a machine learning-based noise suppression, and outputting the noise-suppressed audio output signal.

Abstract: A method for automatic for sound recognition, comprising a) raw spectrogram generation from a sound signal spectrum; b) wide-band spectrum determination; c) wide-band continuous spectrum determination; d) tonal and time-transient spectrum determination; wide-band continuous spectrogram and tonal and time-transient spectrogram determination; and) spectrogram image generation.

Abstract: An acoustic inspection apparatus includes a vibration sound source, a microphone group, and a processor. The vibration sound source emits a vibration sound to an inspection target object. The microphone group includes a first microphone arranged near the inspection target object and a second microphone arranged to have an interval with respect to the first microphone. The processor calculates a first impulse response between the first and second microphones, denoises a component corresponding to the vibration sound from the first impulse response, converts, into a frequency characteristic, a second impulse response obtained from the first impulse response, calculates acoustic energy between the first and second microphones based on the frequency characteristic, and determines an abnormal state of the inspection target object based on the acoustic energy.

Abstract: A memory system includes a non-volatile memory and a memory controller. The memory controller is configured to read a received word from the non-volatile memory, estimate noise by using a plurality of different models for estimating the noise included in the received word to obtain a plurality of noise estimation values, select one noise estimation value from the plurality of noise estimation values, update the received word by using a value obtained by subtracting the selected noise estimation value from the read received word, and decode the updated received word by using a belief-propagation method.

Abstract: According to certain embodiments, an electronic device comprises a microphone configured to acquire a signal including a voice signal and noise signal; a speaker; a memory; and a processor, wherein the processor is configured to: receive the signal from the microphone, wherein the signal corresponds to a plurality of predetermined frequency bands; identify portions of the signal corresponding to a first band and a second band of the plurality of frequency bands; calculate a signal-to-noise ratio (SNR) values for each predetermined frequency band, based on the signal; obtain a first parameter for correcting the portion of the signal corresponding to the first band and a second parameter for correcting the portion of the signal corresponding to the second band, based on the calculated SNR values for the first band and the second band; and apply the first parameter and the second parameter to each of the predetermined frequency bands.

Abstract: The present invention relates to transposing signals in time and/or frequency and in particular to coding of audio signals. More particular, the present invention relates to high frequency reconstruction (HFR) methods including a frequency domain harmonic transposer. A method and system for generating a transposed output signal from an input signal using a transposition factor T is described. The system comprises an analysis window of length La, extracting a frame of the input signal, and an analysis transformation unit of order M transforming the samples into M complex coefficients. M is a function of the transposition factor T. The system further comprises a nonlinear processing unit altering the phase of the complex coefficients by using the transposition factor T, a synthesis transformation unit of order M transforming the altered coefficients into M altered samples, and a synthesis window of length Ls, generating a frame of the output signal.

Abstract: Secure analytics using homomorphic and injective format-preserving encryption are disclosed herein. An example method includes encoding an analytic parameter set using a homomorphic encryption scheme as a set of homomorphic analytic vectors; transmitting the set of homomorphic analytic vectors to a server system; and receiving a homomorphic encrypted result from the server system, the server system having utilized the homomorphic encryption scheme and a first injective, format-preserving encryption scheme to evaluate the set of homomorphic analytic vectors over a datasource.

Abstract: An electronic device and a method for controlling the electronic device are disclosed. The electronic device of the disclosure includes a microphone, a memory storing at least one instruction, and a processor configured to execute the at least one instruction. The processor, by executing the at least one instruction, is configured to: obtain second voice data by inputting first voice data input via the microphone to a first model trained to enhance sound quality, obtain a weight by inputting the first voice data and the second voice data to a second model, and identify input data to be input to a third model using the weight.

Abstract: A system configured to perform sound source localization (SSL) using reflection classification is provided. A device processes audio data representing sounds from multiple sound sources to generate sound track data that includes an individual sound track for each of the sound sources. To detect reflections, the device determines whether a pair of sound tracks are strongly correlated. For example, the device may calculate a correlation value for each pairwise combination of the sound tracks and determine whether the correlation value exceeds a threshold value. When the correlation value exceeds the threshold, the device invokes a reflection classifier trained to distinguish between direct sound sources and reflected sound sources. For example, the device extracts feature data from the pair of sound tracks and processes the feature data using a trained model to determine which of the sound tracks corresponds to the direct sound source.

Abstract: Methods, systems, and devices for signal processing are described. Generally, as provided for by the described techniques, a wearable device to receive an input audio signal from one or more outer microphones, an input audio signal from one or more inner microphones, and a bone conduction signal from a bone conduction sensor based on the input audio signals. The wearable device may filter the bone conduction signal based on a set of frequencies of the input audio signals, such as a low frequency portion of the input audio signals. For example, the wearable device may apply a filter to the bone conduction signal that accounts for an error in the input audio signals. The wearable device may add a gain to the filtered bone conduction signal and may equalize the filtered bone conduction signal based on the gain. The wearable device may output an audio signal to a speaker.

Abstract: A morphing interface system updates, that is, morphs, a display on a client device as a user provides portions of input. The system receives a first portion of user input and selects an intent related to a predicted intent of the user that matches or can be associated with the portion of user input. The system displays an interface associated with the intent at the client device. The system additionally extracts entity values associated with execution of the intent from the first portion of the user input. As the user provides additional portions of input, the morphing interface system reevaluates the user input and determines whether the user input still indicates the same user intent, and thus the same intent for execution. Upon determining that the additional input indicates a different user intent, the system updates the display to show an interface associated with the new intent.

Abstract: A hearing device adapted for being located at or in an ear of a user, or for being fully or partially implanted in the head of a user comprises a) an input unit for providing at least one electric input signal representing sound in an environment of the user, said electric input signal comprising a target speech signal from a target sound source and additional signal components, termed noise signal components, from one or more other sound sources, b) a noise reduction system for providing an estimate of said target speech signal, wherein said noise signal components are at least partially attenuated, and c) an own voice detector for repeatedly estimating whether or not, or with what probability, said at least one electric input signal, or a signal derived therefrom, comprises speech originating from the voice of the user.

Abstract: A likelihood of voice that is a discrimination measure between noise and voice is corrected, by using a Gaussian mixture model of noise learned in a time section in which an input signal is noise, and a voice activity is detected on the basis of the corrected likelihood of voice.

Abstract: A named entity recognition method, a named entity recognition equipment and a medium are disclosed, the method including: acquiring a voice signal; extracting a voice feature vector in the voice signal; extracting, based on a literalness result after voice recognition is performed on the voice signal, a literalness feature vector in the literalness result; splicing the voice feature vector and the literalness feature vector to obtain a composite feature vector of each word in the voice signal; processing the composite feature vector of each word in the voice signal through a deep learning model to obtain a named entity recognition result.

Abstract: A voiceover device is disclosed for use with an audio-visual device having a video display and an audio output. The voiceover device comprises an input device for receiving voiceover audio information from a user. A receiver receives the voiceover audio information of the user. A voiceover circuit is connected to the receiver for amplifying the voiceover audio information for driving a speaker. A muting circuit is connected to the receiver for muting the audio output of the audio-visual device upon the receiver receiving a voiceover audio input from the input device to substitute the voiceover audio information from the user for the original audio output of the audio-video display.

Abstract: Described herein is a system for responding to a frustrated user with a response determined based on spoken language understanding (SLU) processing of a user input. The system detects user frustration and responds to a repeated user input by confirming an action to be performed or presenting an alternative action, instead of performing the action responsive to the user input. The system also detects poor audio quality of the captured user input, and responds by requesting the user to repeat the user input. The system processes sentiment data and signal quality data to respond to user inputs.

Abstract: An approach for enhancing speech reproduction based on recognizing text segments from sound segments of an audio signal. Enhanced text segments are generated from any text segments whose quality indicators do not reach a threshold level, and enhanced speech segments are synthetized from the enhanced text segments. An enhanced sound is reproduced comprising enhanced sound segments based on the enhanced speech segments.

Abstract: There is provided mechanisms for determining average total transmission power for an antenna array configured for beamformed transmission within an angular coverage region. A method is performed by a control device. The method comprises determining bin-wise values of beamforming gain for a set of non-overlapping bins collectively covering the angular coverage region of the antenna array. The method comprises obtaining values of total transmission power of the beamformed transmission. The method comprises determining bin-wise values of average transmission power from the bin-wise values of beamforming gain and the values of total transmission power. The method comprises combining the bin-wise values of average transmission power into one value of average total transmission power for the antenna array.

Abstract: Invoking one or more previously dormant functions of an automated assistant in response to detecting, based on processing of vision data from one or more vision components: (1) a particular gesture (e.g., of one or more “invocation gestures”) of a user; and/or (2) detecting that a gaze of the user is directed at an assistant device that provides an automated assistant interface (graphical and/or audible) of the automated assistant. For example, the previously dormant function(s) can be invoked in response to detecting the particular gesture, detecting that the gaze of the user is directed at an assistant device for at least a threshold amount of time, and optionally that the particular gesture and the directed gaze of the user co-occur or occur within a threshold temporal proximity of one another.

Abstract: An artificial intelligence device includes a microphone configured to receive a voice command of a user, a communication unit configured to wirelessly perform communication with a first slave artificial intelligence device and a second slave artificial intelligence device, and a processor configured to acquire an intention of the voice command and determine a device which will perform operation corresponding to the intention based on the acquired intention and information on each of the artificial intelligence device, the first slave artificial intelligence device and the second slave artificial intelligence device.

Abstract: Implementations can reduce the time required to obtain responses from an automated assistant by, for example, obviating the need to provide an explicit invocation to the automated assistant, such as by saying a hot-word/phrase or performing a specific user input, prior to speaking a command or query. In addition, the automated assistant can optionally receive, understand, and/or respond to the command or query without communicating with a server, thereby further reducing the time in which a response can be provided. Implementations only selectively initiate on-device speech recognition responsive to determining one or more condition(s) are satisfied. Further, in some implementations, on-device NLU, on-device fulfillment, and/or resulting execution occur only responsive to determining, based on recognized text form the on-device speech recognition, that such further processing should occur.

Abstract: Audio systems, methods, and processor instructions are provided that detect voice activity of a user and provide an output voice signal. The systems, methods, and instructions receive a plurality of microphone signals and combine the plurality of microphone signals according to a first combination and a second combination. The first combination produces a primary signal having enhanced response in the direction of the user's mouth, and the second combination produces a reference signal having reduced response in the direction of the user's mouth. The primary signal and the reference signal are added and subtracted to produce a summation signal and a difference signal, respectively. The summation signal and the difference signal are compares and an output voice signal is provided based upon the comparison.

Abstract: A system automatically frames locations by detecting a user's presence within a virtual detection space. The system detects sound in the detection space and converts the sound into electrical signals. The electrical signals are converted into a digital signals at common or periodic sampling rates. The system identifies speech segments in the digital signals and attenuates noise like components within or adjacent to them. The system identifies the physical locations of a speech source generating the speech segments and automatically adjusts a camera framing based on the estimated location of the active speech source generating the speech segments.

Abstract: A method of determining a highlight section of a sound source includes obtaining a sound source and classification information of the sound source, and learning a neural network by using the sound source and the classification information. The neural network includes an input layer including a node corresponding to a feature value of each of a plurality of sections obtained by splitting the sound source according to a time axis, an output layer including a node corresponding to the classification information, a hidden layer defined between the input layer and the output layer, a first function between the input layer and the hidden layer, and a second function between the hidden layer and the output layer, wherein the first function includes an attention model for calculating a weighted sum of the feature value of each section. The highlight section of the sound source is determined based on weight information of a feature value node of each section included in the first function.

Abstract: A method comprising determining an envelope of an audio file based on a double-windowing analysis of the audio file.

Abstract: Hearing aids for persons with sensorineural hearing loss aim to compensate for degraded speech perception caused by frequency-dependent elevation of hearing thresholds, reduced dynamic range, abnormal loudness growth, and increased temporal and spectral masking. A digital hearing aid is implemented as a smartphone application as an alternative to ASIC-based hearing aids. The implementation provides user-configurable processing for background noise suppression and dynamic range compression. Both processing blocks are implemented for real-time processing using single FFT-based analysis-synthesis. A touch-controlled graphical user interface enables the user to set and fine-tune the processing parameters in an interactive and real-time mode.

Abstract: Audio pickup systems and methods are provided to enhance an audio signal by removing noise components related to an acoustic environment. The systems and methods receive a primary signal and one or more reference signals from various microphones. Adaptive filtering and combining minimizes an energy content of a resulting output signal, e.g., to form a substantially null output when the system is in a static acoustic environment. When the system is a playback sound source, one or more echo cancellers may contribute to removing content from the output signal. A change in the acoustic environment, such as a new sound source, causes content in the output signal until the adaptive filtering adapts to the new environment. In some examples, a desired content such as a wake-up word is detected and adaptation is stopped.

Abstract: A method for detecting an audio signal and an apparatus, where the method includes determining a segmental signal-to-noise ratio (SSNR) of an audio signal in response to the audio signal being an unvoiced signal, reducing a reference voice activity detection (VAD) decision threshold to obtain a reduced VAD decision threshold, and comparing the SSNR with the reduced VAD decision threshold to determine whether the audio signal is an active signal.

Abstract: A computer includes a processor and a memory storing instructions executable by the processor to receive sensor data of an occupant of a vehicle, identify a type of mask worn by the occupant based on the sensor data, select a sound filter according to the type of mask from a plurality of sound filters stored in the memory, receive sound data, apply the selected sound filter to the sound data, and perform an operation using the filtered sound data.

Abstract: Techniques for improving microphone noise suppression are provided. A system for noise-suppression may include a beam selector component that applies logic to select a beam most likely corresponding to a direction of a noise source and keeps the beam selection steady rather than switching the beam too often to avoid processing complications. The selected beam may be used as a reference in an adaptive filter which outputs a noise estimate. The noise estimate and raw microphone data may be used to adapt the adaptive filter. A parallel filter which adapts after a time delay may be applied to the reference in order to prevent interference. An attenuation factor may be used to scale the noise estimate based on noise diffuseness, signal quality, and/or a gain limit. The scaled noise estimate may be subtracted from microphone input data to produce output audio data with improved signal quality and maintained signal coherence.

Abstract: A process for compressing an audio speech signal utilizes ASR processing to generate a corresponding text representation and, depending on confidence in the corresponding text representation, selectively applies more, less, or no compression to the audio signal. The result is a compressed audio signal, with corresponding text, that is compact and well suited for searching, analytics, or additional ASR processing.

Abstract: A method of enhancing distorted signal having a speech and noise component, with a processing device having memory with stored training information T, comprising a step of removing noise in spectrum domain according to a noise and speech model to obtain a clear signal spectrum, wherein the training information T comprises dictionaries of compact spectra prototypes of speech and noise, speech gains and noise gains forming together composite states and it comprises probabilities of state history. Dictionaries of compact spectra prototypes comprise compact spectra prototypes of a first dimension P1 and of a second dimension P2 where second dimension P2 is higher or equal to the first dimension P1.

Abstract: A method for multi-channel voice activity detection includes receiving a sequence of input frames characterizing streaming multi-channel audio captured by an array of microphones. Each channel of the streaming multi-channel audio includes respective audio features captured by a separate dedicated microphone. The method also includes determining, using a location fingerprint model, a location fingerprint indicating a location of a source of the multi-channel audio relative to the user device based on the respective audio features of each channel of the multi-channel audio. The method also includes generating an output from an application-specific classifier. The first score indicates a likelihood that the multi-channel audio corresponds to a particular audio type that the particular application is configured to process.

Abstract: Systems and methods for processing an audio signal include an audio input operable to receive an input signal comprising a time-domain, single-channel audio signal, a subband analysis block operable to transform the input signal to a frequency domain input signal comprising a plurality of k-spaced under-sampled subband signals, a reverberation reduction block operable to reduce reverberation effect, including late reverberation, in the plurality of k-spaced under-sampled subband signals, a noise reduction block operable to reduce background noise from the plurality of k-spaced under-sampled subband signals, and a subband synthesis block operable to transform the subband signals to the time-domain, thereby producing an enhanced output signal.

Abstract: Operations related to performing gain operations with respect to a receive-path signal of a first device may be performed. The operations may include obtaining the receive-path signal, which includes an echo speech signal and a receive speech signal originating at a second device. In addition, the operations may include identifying a portion of the receive-path signal that includes, at a particular time, a first frequency component that corresponds to the echo speech signal and a second frequency component that corresponds to the receive speech signal in which the first frequency component is different from the second frequency component. Moreover, the operations may include attenuating the first frequency component of the portion while avoiding attenuating the second frequency component of the portion based on the first frequency component corresponding to the echo speech signal and the second frequency component corresponding to the receive speech signal.

Abstract: Aspects of the present disclosure relate to voice command filtering. Communication is established with a voice controlled device located at a location. Data indicating blocked directions is received from the voice controlled device. A voice command is received. A determination is made that the voice command is received from a blocked direction indicated in the data. The received voice command is then ignored.

Abstract: A system is provided. The system comprises at least one artificial neural network configured to: receive an audio signal; for a time period, determine if at least one human voice audio spectrum is in the audio signal; for the time period, identify at least one human voice audio power spectrum; for the time period, extract each of the at least one identified human voice audio power spectrum; remove artifacts from each extracted human voice audio power spectrum to synthesize an estimation of an original human voice prior to its distortion; and transmit the synthesized estimation of an original human voice.

Abstract: In a general aspect, an apparatus for detecting speech can include a signal conditioning stage that receives a signal corresponding with acoustic energy, filters the received signal to produce a speech-band signal, calculates a first sequence of energy values for the received signal and calculates a second sequence of energy values for the speech-band signal. The apparatus can also include a detection stage including a plurality of speech and noise differentiators. The detection stage can being configured to receive the first and second sequences of energy values and, based on the first sequence of energy values and the second sequence of energy values, provide, for each speech and noise differentiator of the plurality of speech and noise differentiators, a respective speech-detection indication signal.

Abstract: With recent real-world applications of speaker and speech recognition systems, robust features for degraded speech have become a necessity. In general, degraded speech results in poor performance of any speech-based system. This poor performance can be attributed to feature extraction functionality of speech-based system which takes input speech file and converts it into a representation called as a feature. Embodiments of the present disclosure provide systems and methods that compute distance between each degraded speech feature extracted from an input speech signal with each clean speech feature comprised in a memory of the system to obtain set of matched clean speech features wherein at least a subset of cleaned speech features are dynamically selected based on a pre-defined threshold and the computed distance, thereby computing statistics for the dynamically selected clean speech features set for utilizing in at least one of a speech recognition system and a speaker recognition system.

Abstract: An artificial intelligence (AI) device may acquire a probability that a received speech signal is classified as a noise signal, calculate a confidence level of a first model for determining to which phoneme the speech signal belongs, based on the speech signal, determine a weight of the first model based on the probability and the confidence level of the first model, and output a speech recognition result of the speech signal using the determined weight of the first model.

Abstract: Methods and devices for processing and voice operated control are provided. The method can include performing a non-difference comparison between a first received sound and a second received sound, determining if speech exists based on the comparison, and transmitting or providing a decision that the speech is present to at least one among the device, a cell phone, a media player, or a portable computing device. Other embodiments are disclosed.