Google Inc.

Deepmind Technologies Limited

Sze Chie Lim

Willem Bastiaan Kleijn

Jan K Skoglund

Alejandro Luebs

Oriol Vinyals

Aaron Gerard Antonius Van Den Oord

Thomas Chadwick Walters

Yazhe Li

Cristina Garbacea

Michael Takezo Chinen

Rajeev Conrad Nongpiur

Marco Tagliasacchi

Neil Zeghidour

Andrew Allen

Qian Zhang

Leonardo Kusumo

Hélène Bahu

Jihan Li

Andrew James Sutter

Kung-Wei Liu

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

The technology generally relates to spatial audio communication between devices. For example, a first device and a second device may be connected via a communication link. The first device may capture audio signals in an environment through two or more microphones. The first device may encode the captured audio with spatial configuration data. The first device may transmit the encoded audio via the communication link to the second device. The second device may decode the encoded audio into binaural or ambisonic audio to be output by one or more speakers of the second device. The binaural or ambisonic audio may be converted into spatial audio to be output. The second device may output the binaural or spatial audio to create an immersive listening experience.

Spatial audio recording from home assistant devices

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

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating discrete latent representations of input audio data. Only the discrete latent representation needs to be transmitted from an encoder system to a decoder system in order for the decoder system to be able to effectively to decode, i.e., reconstruct, the input audio data.

Speech coding using content latent embedding vectors and speaker latent embedding vectors

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

A method includes: receiving time instants of audio signals generated by a set of microphones at a location; determining a distortion measure between frequency components of at least some of the received audio signals; determining a similarity measure for the frequency components using the determined distortion measure; and processing the audio signals based on the determined similarity measure.

Blind source separation using similarity measure

Techniques of rendering sound for a listener involve producing, as the amplitude of each of the source driving signals, a sum of two terms: a first term based on a solution sto the equation b=A&#xb7;s, and a second term based on a projection of a specified vector &#x15d; onto the nullspace of A, &#x15d; not being a solution to the equation b=A&#xb7;s. Along these lines, in one example, the first term is equivalent to a Moore-Penrose pseudoinverse, e.g., A(AA)&#xb7;b. In general, any solution to the equation b=A&#xb7;s is satisfactory. The specified vector that is projected onto the nullspace of A is defined to reduce the coherence of the net sound field. Advantageously, the resulting operator is both linear time-invariant and idempotent so that the sound field may be faithfully reproduce both inside the RSF and at a sufficient range outside the RSF to cover a human head.

Incoherent idempotent ambisonics rendering

Growing community of inventors

San Francisco, CA, United States of America

Sze Chie Lim