Disclosed are methods, systems, device, and other implementations, including a method that includes receiving an audio signal representation, detecting in the received audio signal representation, using a first learning model, one or more silent intervals with reduced foreground sound levels, determining based on the detected one or more silent intervals an estimated full noise profile corresponding to the audio signal representation, and generating with a second learning model, based on the received audio signal representation and on the determined estimated full noise profile, a resultant audio signal representation with a reduced noise level.

A controller for an acoustic echo canceller includes a noise estimator configured to estimate a level of noise that is comprised in a microphone signal relative to an echo component, estimated by the acoustic echo canceller, comprised in the microphone signal. The controller further includes a control module configured to control the acoustic echo canceller in dependence on that estimate.

In a method for coding of information for enhancing a background noise representation, voice activity of an input speech signal is determined. A noisiness parameter is determined for an inactive speech signal, wherein the noisiness parameter is based on a ratio of prediction gains of two Linear Predictive Coder (LPC) prediction filters with different orders. The noisiness parameter is quantized, and the quantized noisiness parameter is encoded for transmission.

Techniques are provided for reduction of echo in a received audio signal based on infinite impulse response (IIR) acoustic echo cancellation in the frequency domain. A methodology implementing the techniques according to an embodiment includes estimating an echo path transfer function associated with the received audio signal. The received audio signal includes a combination of a speech signal and a reference signal modified by the echo path transfer function. The estimation employs an IIR filter and a finite impulse response (FIR) filter, both of which operate in the frequency domain. The IIR filter is configured to model longer term echo components and the FIR filter is configured to model shorter term echo components. The method further includes applying the filters to the reference signal to generate an echo correction signal which is subtracted from the received audio signal to reduce the echo and generate an estimate of the speech signal.

A controller for an acoustic echo canceller is described. The controller includes a noise estimator configured to estimate a level of noise that is comprised in a microphone signal relative to an echo component, estimated by the acoustic echo canceller, comprised in the microphone signal. The controller further includes a control module configured to control the acoustic echo canceller in dependence on that estimate.

An audio device comprising an interface, a memory, and one or more processors is disclosed, wherein the one or more processors are configured to: obtain a microphone input signal; obtain a media input signal; process the microphone input signal for provision of a microphone output signal; and provide an audio output signal based on the microphone output signal, wherein to process the microphone input signal comprises to: determine a microphone gain; and apply the microphone gain to the microphone input signal for provision of the microphone output signal, and wherein to determine the microphone gain comprises: estimate a first loudness of the microphone input signal; determine a first primary average based on the first loudness; estimate a second loudness of the media input signal; determine a second average based on the second loudness; determine a first gain based on the first primary average and the second average; determine a first secondary average based on the first loudness; determine a second gain based on the first secondary average and the second average; and determine the microphone gain based on the first gain and the second gain.

An apparatus includes a first calculator configured to determine a long-term noise estimate of the audio signal. The apparatus also includes a second calculator configured to determine a formant-sharpening factor based on the determined long-term noise estimate. The apparatus includes a filter configured to filter a codebook vector to generate a filtered codebook vector. The filter is based on the determined formant-sharpening factor, and the codebook vector is based on information from the audio signal. The apparatus further includes an audio coder configured to generate a formant-sharpened low-band excitation signal based on the filtered codebook vector.

This invention relates to a method of and a noise suppression system (100) for noise suppression of a sound signal, the sound signal comprising speech of a user (110) when the user (110) is speaking, the system comprising at least one first sound receiver (101) adapted to obtain, during use, a first sound signal (120), and at least one second sound receiver (102) adapted to obtain, during use, a second sound signal (121), wherein the first sound signal (120) comprises a first airborne noise signal (103) when noise (111) is present and a first airborne speech signal (104) when the user (110) is speaking, the second sound signal (121) comprises a second airborne noise signal (105) when noise (111) is present and a second airborne speech signal (106) when the user (110) is speaking, the at least one first sound receiver (101) is a vibration pickup or transducer (101) adapted to obtain, during use, an additional speech signal (107) when the user (110) is speaking, wherein the additional speech signal (107) is obtained directly or indirectly in response to vibrations propagating through the user (110), the vibrations being caused by the user (110) speaking, and the first sound signal (120) further comprises the additional speech signal (107) when the user (110) is speaking, wherein the system (100) is adapted to suppress, during use, at least a part of the first airborne noise signal (103), when present, in the first sound signal (120).

FIG. 2 is to be published.

Methods, systems, and computer-readable storage media for voice denoising. Implementations include actions of performing a mathematical transform on each frame signal in an audio signal segment to generate multiple power spectra. Each power spectrum corresponds to a respective frame signal. Power value variances corresponding to frame signals at various frequencies are determined. A noise signal is identified in each frame signal based on the power value variance. The identified noise signal is removed from each frame signal of the plurality of frame signals.

A comfort noise generation apparatus constituted of: near and far end speech detectors arranged to detect speech activity in near-end and far-end signals and a comfort noise generator, wherein responsive to an indication from the near-end speech detector that speech activity is absent on the near-end signal and an indication from the far-end silence detector that speech activity is absent on the far-end signal, the comfort noise generator is arranged to initiate a determination of an estimation of near-end background noise, wherein responsive to an indication from the near-end speech detector that speech activity is present on the near-end signal or an indication from the far-end silence detector that speech activity is present on the far-end signal, the comfort noise generator is arranged to terminate the estimation determination of near-end background noise, and wherein the comfort noise generator is arranged to output a function of the near-end background noise estimation.