Abstract: Embodiments and methods perform ear cavity frequency response (EFCR) adaptive noise cancelation (ANC) with path-compensation over an entire main path to the eardrum (MPED) of a user. A number of ANC filter models are pre-trained to include respective anti-noise path (ANP) filter models and respective MPED filter models representing ANC filter configurations. As a user wears a headphone earpiece, characteristics of the wearer and the position/orientation of wearing manifest a wearer/wearing condition. Techniques described herein can continuously or periodically and efficiently determine which of the pre-trained ANC filter models most closely described the present MPED of the present wearer/wearing condition, and can continuously or periodically update the ANC filter configuration based on the pre-trained models to maintain high-performance ANC that includes EFCR path-compensation.

Abstract: Embodiments and methods perform ear cavity frequency response (EFCR) adaptive noise cancelation (ANC) with path-compensation over an entire main path to the eardrum (MPED) of a user. A number of ANC filter models are pre-trained to include respective anti-noise path (ANP) filter models and respective MPED filter models representing ANC filter configurations. As a user wears a headphone earpiece, characteristics of the wearer and the position/orientation of wearing manifest a wearer/wearing condition. Techniques described herein can continuously or periodically and efficiently determine which of the pre-trained ANC filter models most closely described the present MPED of the present wearer/wearing condition, and can continuously or periodically update the ANC filter configuration based on the pre-trained models to maintain high-performance ANC that includes EFCR path-compensation.

Abstract: Pre-processing systems, methods of pre-processing, and speech processing systems for improved Automated Speech Recognition are provided. Some pre-processing systems for improved speech recognition of a speech signal are provided, which systems comprise a pitch estimation circuit; and a pitch equalization processor. The pitch estimation circuit is configured to receive the speech signal to determine a pitch index of the speech signal, and the pitch equalization processor is configured to receive the speech signal and pitch information, to equalize a speech pitch of the speech signal using the pitch information, and to provide a pitch-equalized speech signal.

Abstract: An apparatus includes a hybrid adaptive active noise control unit (HAANCU) configured to provide an anti-noise signal to an ear speaker from a reference noise signal of a reference microphone and an error signal of an error microphone, a decimator configured to decimate the reference noise signal and error signal, an adaptive hybrid ANC training unit (AHANCTU) including at least one noise cancellation filter and a filter configured to provide a feedback signal to the at least one noise cancellation, which trains parameters of the AHANCTU based on the decimated reference noise signal, the decimated error signal, and the feedback signal. The apparatus further includes a rate conversion unit configured to up-sample the parameters and update the HAANCU with the up-sampled parameters.

Abstract: An apparatus for canceling noise at an ear speaker includes a wideband active noise cancellation filter having a first bandwidth and configured to generate a wideband anti-noise signal from a received reference noise signal, a narrowband active noise cancellation filter having a second bandwidth smaller than the first bandwidth and configured to generate a narrowband anti-noise signal from an error noise signal, a filter between the ear speaker and an error microphone and configured to generate a feedback noise signal, and a controller. The controller is configured to eliminate the error noise signal by modifying coefficients of the wideband active noise cancellation filter and the narrowband active noise cancellation filter in response to the wideband anti-noise signal, the narrowband anti-noise signal, and the feedback noise signal.

Abstract: An apparatus includes a hybrid adaptive active noise control unit (HAANCU) configured to provide an anti-noise signal to an ear speaker from a reference noise signal of a reference microphone and an error signal of an error microphone, a decimator configured to decimate the reference noise signal and error signal, an adaptive hybrid ANC training unit (AHANCTU) including at least one noise cancellation filter and a filter configured to provide a feedback signal to the at least one noise cancellation, which trains parameters of the AHANCTU based on the decimated reference noise signal, the decimated error signal, and the feedback signal. The apparatus further includes a rate conversion unit configured to up-sample the parameters and update the HAANCU with the up-sampled parameters.

Abstract: An apparatus for canceling noise at an ear speaker includes a wideband active noise cancellation filter having a first bandwidth and configured to generate a wideband anti-noise signal from a received reference noise signal, a narrowband active noise cancellation filter having a second bandwidth smaller than the first bandwidth and configured to generate a narrowband anti-noise signal from an error noise signal, a filter between the ear speaker and an error microphone and configured to generate a feedback noise signal, and a controller. The controller is configured to eliminate the error noise signal by modifying coefficients of the wideband active noise cancellation filter and the narrowband active noise cancellation filter in response to the wideband anti-noise signal, the narrowband anti-noise signal, and the feedback noise signal.

Abstract: An apparatus includes a hybrid adaptive active noise control unit (HAANCU) configured to provide an anti-noise signal to an ear speaker from a reference noise signal of a reference microphone and an error signal of an error microphone, a decimator configured to decimate the reference noise signal and error signal, an adaptive hybrid ANC training unit (AHANCTU) including at least one noise cancellation filter and a filter configured to provide a feedback signal to the at least one noise cancellation, which trains parameters of the AHANCTU based on the decimated reference noise signal, the decimated error signal, and the feedback signal. The apparatus further includes a rate conversion unit configured to up-sample the parameters and update the HAANCU with the up-sampled parameters.

Abstract: Pre-processing systems, methods of pre-processing, and speech processing systems for improved Automated Speech Recognition are provided. Some pre-processing systems for improved speech recognition of a speech signal are provided, which systems comprise a pitch estimation circuit; and a pitch equalization processor. The pitch estimation circuit is configured to receive the speech signal to determine a pitch index of the speech signal, and the pitch equalization processor is configured to receive the speech signal and pitch information, to equalize a speech pitch of the speech signal using the pitch information, and to provide a pitch-equalized speech signal.

Abstract: A method, a device, and a non-transitory storage medium are described in which a power of late reverberation of a speech signal is estimated based on early samples of the speech signal. The power of the late reverberation may be subtracted linearly or non-linearly from the speech signal.

Abstract: A method, a device, and a non-transitory storage medium are described in which a power of late reverberation of a speech signal is estimated based on early samples of the speech signal. The power of the late reverberation may be subtracted linearly or non-linearly from the speech signal.

Abstract: Techniques, described herein, may enable a computing device (e.g., a set-top-box, a user equipment (UE), etc.) with an array of sensors (e.g., microphones, antennas, etc.) to engage in beamforming in a manner that optimizes the signal strength of the target signal. For example, the computing device may create an optimized beam pattern by applying a pre-selected band width to a steering direction of the target signal. The beam width may include steering vectors at equal, angular distances from the steering direction, enabling steep transition slopes to be applied to the optimized beam pattern. In some implementations, the beam width and transition slopes may be standardized, such that the optimized beam pattern for any signal may be uniform, regardless of variables such as frequency and time.

Abstract: Techniques, described herein, may enable a computing device (e.g., a set-top-box, a user equipment (UE), etc.) with an array of sensors (e.g., microphones, antennas, etc.) to engage in beamforming in a manner that optimizes the signal strength of the target signal. For example, the computing device may create an optimized beam pattern by applying a pre-selected band width to a steering direction of the target signal. The beam width may include steering vectors at equal, angular distances from the steering direction, enabling steep transition slopes to be applied to the optimized beam pattern. In some implementations, the beam width and transition slopes may be standardized, such that the optimized beam pattern for any signal may be uniform, regardless of variables such as frequency and time.

Abstract: Examples of the disclosure describe user environment aware single channel acoustic noise reduction. A noisy signal received by a computing device is transformed and feature vectors of the received noisy signal are determined. The computing device accesses classification data corresponding to a plurality of user environments. The classification data for each user environment has associated therewith a noise model. A comparison is performed between the determined feature vectors and the accessed classification data to identify a current user environment. A noise level, a speech level, and a speech presence probability from the transformed noisy signal are estimated and the noise signal is reduced based on the estimates. The resulting signal is outputted as an enhanced signal with a reduced or eliminated noise signal.

Abstract: Architectures of numbers of microphones and their positioning in a device for sound source direction estimation and source separation are presented. The directions of sources are front, back, left, right, top, and bottom of the device, and can be determined by amplitude and phase differences of microphone signals with proper microphone positioning. The source separation is to separate the sound coming from different directions from the mix of sources in microphone signals. This can be done with blind source separation (BSS), independent component analysis (ICA), and beamforming (BF) technologies. The device can perform many kinds of audio enhancements for the device. For example, it can perform noise reduction for communications; it can choose a source from a desired direction to perform speech recognition; and it can correct sound perceiving directions in microphones and generate desired sound images like stereo audio output. In addition, with source separation, 2.1, 5.1, 7.

Abstract: Presented is a method and associated system for suppression of linear and nonlinear echo. The method includes dividing an input signal into several frequency bands in each of a several of time frames. The input signal may include an echo signal. The method further includes multiplying the input signal in each of the several frequency bands by a corresponding echo suppression signal. Calculating the corresponding echo suppression signal may include estimating a power of the echo signal in a particular frequency band as a sum of several component echo powers, each of the several component echo powers due to an excitation from a far-end signal in a corresponding one of the several frequency bands. Calculating the corresponding echo suppression signal may further include subtracting the power of the echo signal in the particular frequency band from a power of the input signal in the particular frequency band.

Abstract: Architectures of numbers of microphones and their positioning in a device for sound source direction estimation and source separation are presented. The directions of sources are front, back, left, right, top, and bottom of the device, and can be determined by amplitude and phase differences of microphone signals with proper microphone positioning. The source separation is to separate the sound coming from different directions from the mix of sources in microphone signals. This can be done with blind source separation (BSS), independent component analysis (ICA), and beamforming (BF) technologies. The device can perform many kinds of audio enhancements for the device. For example, it can perform noise reduction for communications; it can choose a source from a desired direction to perform speech recognition; and it can correct sound perceiving directions in microphones and generate desired sound images like stereo audio output. In addition, with source separation, 2.1, 5.1, 7.

Abstract: A system for detecting motion comprising a first speaker, a first microphone separated from the first speaker by a distance D1, a sound generator, an echo parameter measurement device and an echo parameter monitor, wherein the echo parameter monitor stores two or more sequential echo parameters and generates a motion indicator if the two or more sequential echo parameters indicate a change in an acoustic echo path that exceeds a predetermined threshold.

Abstract: Traditionally, echo cancellation has employed linear adaptive filters to cancel echoes in a two way communication system. The rate of adaptation is often dynamic and varies over time. Disclosed are novel rates of adaptation that perform well in the presence of background noise, during double talk and with echo path changes. Additionally, the echo or residual echo can further be suppressed with non-linear processing performed using joint frequency-time domain processing.

Abstract: A method for continuously estimating reverberation decay comprising receiving a sequence of audio data samples. Determining whether a plateau is present in the sequence of audio data samples. Generating one or more reverberation parameters from the sequence of audio data samples if it is determined that the plateau is present.