Abstract: 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.

Abstract: 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.

Abstract: A method including generating an audio stream including a first substream as first audio data and a second substream as second audio data, generating a first loudness parameter associated with playback of the first substream, generating a second loudness parameter associated with playback of the second substream, and generating an audio package including an identification corresponding to the first audio data, an identification corresponding to the second audio data, and a codec agnostic container including the first loudness parameter, and the second loudness parameter.

Abstract: A method including receiving first audio data, receiving second audio data, compressing the first audio data as first compressed audio data, compressing the second audio data as second compressed audio data, generating a codec dependent container including a parameter associated with compressing the first audio data, compressing the second audio data, a reference to the first compressed audio data, and a reference to the second compressed audio data, generating a codec agnostic container including at least one parameter representing time-varying data associated with playback of the first audio data and the second audio data, and generating an audio package including the codec dependent container and the codec agnostic container.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.

Abstract: 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.