Google Inc.

Global IP Solutions, Inc.

Global IP Solutions (Gips) Ab

Global IP Sound Ab

Jan K Skoglund

Willem Bastiaan Kleijn

Sze Chie Lim

Andrew John MacDonald

Alejandro Luebs

Minyue Li

Simon J Godsill

Björn Volcker

Herbert Buchner

Andrew Hines

Anil Christopher Kokaram

Turaj Zakizadeh Shabestary

Patrick A Naylor

Marco Paniconi

Yiteng Huang

Andrew Allen

Koen Bernard Vos

Roar Hagen

Niklas Enbom

Michael Takezo Chinen

Jan T Linden

Naomi Harte

Jens Enzo Nyby Christensen

Noami A Harte

Alastair H Moore

D James Eaton

Miroslaw Narbutt

Christos Tzagkarakis

Jan Thomas Linden

Koen Vos

John Andrew Macdonald

Implementations identify a small set of independent, salient features from an input signal. The salient features may be used for conditioning a generative network, making the generative network robust to noise. The salient features may facilitate compression and data transmission. An example method includes receiving an input signal and extracting salient features for the input signal by providing the input signal to an encoder trained to extract salient features. The salient features may be independent and have a sparse distribution. The encoder may be configured to generate almost identical features from two input signals a system designer deems equivalent. The method also includes conditioning a generative network using the salient features. In some implementations, the method may also include extracting a plurality of time sequences from the input signal and extracting the salient features for each time sequence.

Identifying salient features for generative networks

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for coding speech using neural networks. One of the methods includes obtaining a bitstream of parametric coder parameters characterizing spoken speech; generating, from the parametric coder parameters, a conditioning sequence; generating a reconstruction of the spoken speech that includes a respective speech sample at each of a plurality of decoder time steps, comprising, at each decoder time step: processing a current reconstruction sequence using an auto-regressive generative neural network, wherein the auto-regressive generative neural network is configured to process the current reconstruction to compute a score distribution over possible speech sample values, and wherein the processing comprises conditioning the auto-regressive generative neural network on at least a portion of the conditioning sequence; and sampling a speech sample from the possible speech sample values.

Speech coding using auto-regressive generative neural networks

Provided are methods and systems for enhancing speech when corrupted by transient noise (e.g., keyboard typing noise). The methods and systems utilize a reference microphone input signal for the transient noise in a signal restoration process used for the voice part of the signal. A robust Bayesian statistical model is used to regress the voice microphone on the reference microphone, which allows for direct inference about the desired voice signal while marginalizing the unwanted power spectral values of the voice and transient noise. Also provided is a straightforward and efficient Expectation-maximization (EM) procedure for fast enhancement of the corrupted signal. The methods and systems are designed to operate easily in real-time on standard hardware, and have very low latency so that there is no irritating delay in speaker response.

Detection and suppression of keyboard transient noise in audio streams with aux keybed microphone

Provided are methods, systems, and apparatus for hierarchical decorrelation of multichannel audio. A hierarchical decorrelation algorithm is designed to adapt to possibly changing characteristics of an input signal, and also preserves the energy of the original signal. The algorithm is invertible in that the original signal can be retrieved if needed. Furthermore, the proposed algorithm decomposes the decorrelation process into multiple low-complexity steps. The contribution of these steps is generally in a decreasing order, and thus the complexity of the algorithm can be scaled.

Hierarchical decorrelation of multichannel audio

Techniques of source localization and acquisition involve a wideband joint acoustic source localization and acquisition approach in light of sparse optimization framework based on an orthogonal matching pursuit-based grid-shift procedure. Along these lines, a specific grid structure is constructed with the same number of grid points as compared to the on-grid case, but which is 'shifted' across the acoustic scene. More specifically, it is expected that each source will be located close to a grid point in at least one of the set of shifted grids. The sparse solutions corresponding to the set of shifted grids are combined to obtain the source location estimates. The estimated source positions are used as side information to obtain the original source signals.

Joint wideband source localization and acquisition based on a grid-shift approach

Techniques of performing linear acoustic echo cancellation performing a phase correction operation on the estimate of the echo signal based on a clock drift between a capture of an input microphone signal and a playout of a loudspeaker signal. Along these lines, the existence of the clock drift, i.e., a small difference in the sampling rates of the input microphone signal and the loudspeaker signal, can cause processing circuitry in a device configured to perform LAEC operations to generate a filter based on the magnitudes of the short-term Fourier transforms (STFTs) of the input microphone signal and the loudspeaker signal. Such a filter is real-valued and results in a positive estimate of the acoustic echo signal included in the input microphone signal. The phase of this estimate may then be aligned with the phase of the input microphone signal.

Echo cancellation for keyword spotting

A method includes: receiving a representation of a soundfield, the representation characterizing the soundfield around a point in space; decomposing the received representation into independent signals; and encoding the independent signals, wherein a quantization noise for any of the independent signals has a common spatial profile with the independent signal.

Coding of a soundfield representation

Detection and suppression of keyboard transient noise in audio streams with auxiliary keybed microphone

A computing device includes a processor and a memory. The processor is configured to generate spectrograms, for example, using short-time Fourier transform, for a plurality of channels of reference and test ambisonic signals. In some implementations, the test ambisonic signal may be generated by decoding an encoded version of the reference ambisonic signal. The processor is further configured to compare, for each of the plurality of channels of a reference ambisonic signal, at least a patch associated with a channel of the reference ambisonic signal with at least a corresponding patch of a corresponding channel of the test ambisonic signal and determine a localization accuracy of the test ambisonic signal based on the comparison. In some implementations, the comparing may be based on phaseograms of the reference and test ambisonic signals.

Objective quality metrics for ambisonic spatial audio

Growing community of inventors

San Francisco, CA, United States of America

Jan K Skoglund