A method includes receiving a microphone audio signal and a playout audio signal, and determining a frequency representation of the microphone audio signal and a frequency representation of the playout audio signal. For each frequency representation, the method also includes determining features based on the frequency representation. Each feature corresponds to a pair of frequencies of the frequency representation and a period of time between the pair of frequencies. The method also includes determining that a match occurs between a first feature based on the frequency representation of the microphone audio signal and a second feature based on the frequency representation of the playout audio signal, and determining that a delay value between the first feature and the second feature corresponds to an echo within the microphone audio signal.

A system and method are provided for dynamic sound mask adjustment. A sound mask is used for obtaining an impulse response measurement that adjusts a generated sound mask dynamically based on real-time ambient noise parameters, while maintaining echo canceller calibration performance. The system includes a dynamic sound mask generator that includes a noise accumulator and monitor that includes a processor and memory including instructions executed by the processor for performing the dynamic sound mask adjustment. If the sound mask is not in the hysteresis range, the current sound mask level and iteration update rate are adjusted. if the sound mask is in the hysteresis range, the current sound mask level and iteration update rate are maintained.

A playback device is configured to receive, via a network interface, a source stream of audio including first and second channel streams of audio, and to produce, via respective first and second speaker drivers, a first channel audio output and a second channel audio output. The playback device is also configured to receive, via one or more microphones, a captured stream of audio including first and second portions corresponding to the respective first and second channel audio outputs. The playback device is also configured to combine at least the first channel stream of audio and the second channel stream of audio into a compound audio signal and perform acoustic echo cancellation on the compound audio signal and thereby produce an acoustic echo cancellation output, then to apply the acoustic echo cancellation output to the captured stream of audio and thereby increase a signal-to noise ratio of the captured stream of audio.

A multi-channel acoustic echo cancellation (AEC) system that includes a step-size controller that dynamically determines a step-size value for each channel and each tone index on a frame-by-frame basis. The system determines that near-end signals are present by calculating a scaled error and determining that the scaled error exceeds a threshold value. When the scaled error exceeds the threshold value, the system may switch from a first cost function to a second cost function and determine a step-size value using a robust algorithm. The robust algorithm may prevent the system from diverging due to the presence of the near-end signal. For example, the robust algorithm may select a different cost function to determine the step-size value and/or combine different step-size computations, resulting in the step-size value being temporarily reduced. Thus, the robust algorithm may enable the AEC to better model the near-end disturbance statistics while the near-end signal is present.

A method for performing echo cancellation includes: receiving a far-end signal from a far-end device at a near-end device; recording a microphone signal at the near-end device including: a near-end signal; and an echo signal corresponding to the far-end signal; extracting far-end features from the far-end signal; extracting microphone features from the microphone signal; computing estimated near-end features by supplying the microphone features and the far-end features to an acoustic echo cancellation module including: an echo estimator including a first stack of a recurrent neural network configured to compute estimated echo features based on the far-end features; and a near-end estimator including a second stack of the recurrent neural network configured to compute the estimated near-end features based on an output of the first stack and the microphone signal; computing an estimated near-end signal from the estimated near-end features; and transmitting the estimated near-end signal to the far-end device.

Provided in the present disclosure are a voice processing method, an apparatus, an electronic device, and a storage medium, the method comprising: detecting the working state of a current call system, and when the working state is a two-end speaking state or a remote-end speaking state, performing compression processing on a subsequent remote-end voice signal, acquiring a near-end voice signal by means of a microphone, performing echo processing on the basis of the near-end voice signal and the compression-processed remote-end voice signal to obtain an echo-processed near-end voice signal and a remaining echo signal, performing non-linear suppression processing on the near-end voice signal and the remaining echo signal, and performing gain control on the suppression-processed near-end voice signal.

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.

Some implementations involve receiving a content stream that includes audio data, receiving at least one type of level adjustment indication relating to playback of the audio data and controlling a level of the input audio data, based on the at least one type of level adjustment indication, to produce level-adjusted audio data. Some examples involve determining, based at least in part on the type(s) of level adjustment indication, a multiband limiter configuration, applying the multiband limiter to the level-adjusted audio data, to produce multiband limited audio data and providing the multiband limited audio data to one or more audio reproduction transducers of an audio environment.

A control unit of a telephone device is configured to perform a first processing of: estimating a line echo amount based on a volume of a DTMF signal generated by a DTMF signal generation unit and a volume of the DTMF signal input from a line; calculating a howling noise threshold value based on the maximum value of the acoustic echo generated at a handset that is measured in advance and set values of a sidetone volume, a transmission volume, and a reception volume; and calculating a difference between the line echo amount and the howling noise threshold value as an adjustment requisite amount and lowering the set value of the sidetone volume by the adjustment requisite amount when the line echo amount is larger than the howling noise threshold value.

An communication system supports communication paths within an environment by receiving speech signals of a speaker and playing it back for one or more listeners. Signal processing tasks are split into a microphone related part and into a loudspeaker related part. A sound processing system suitable for use in an environment having multiple acoustic zones includes a plurality of microphone communication instances coupled and a plurality of loudspeaker instances.