Google Inc.

Qian Zhang

Hasim Sak

Anshuman Tripathi

Han Lu

Jaeyoung Kim

Soheil Khorram

Rajeev Conrad Nongpiur

Khe Chai Sim

Marco Tagliasacchi

Sze Chie Lim

Byungchul Kim

Neil Zeghidour

Leonardo Kusumo

Hélène Bahu

Michael Takezo Chinen

Marie Vachovsky

Monica Song

Kim Jaeyoung

Kung-Wei Liu

Jihan Li

Andrew James Sutter

A method includes receiving a plurality of unlabeled audio samples corresponding to spoken utterances not paired with corresponding transcriptions. At a target branch of a contrastive Siamese network, the method also includes generating a sequence of encoder outputs for the plurality of unlabeled audio samples and modifying time characteristics of the encoder outputs to generate a sequence of target branch outputs. At an augmentation branch of a contrastive Siamese network, the method also includes performing augmentation on the unlabeled audio samples, generating a sequence of augmented encoder outputs for the augmented unlabeled audio samples, and generating predictions of the sequence of target branch outputs generated at the target branch. The method also includes determining an unsupervised loss term based on target branch outputs and predictions of the sequence of target branch outputs. The method also includes updating parameters of the audio encoder based on the unsupervised loss term.

Contrastive Siamese network for semi-supervised speech recognition

A method includes receiving a sequence of acoustic frames extracted from unlabeled audio samples that correspond to spoken utterances not paired with any corresponding transcriptions. The method also includes generating, using a supervised audio encoder, a target higher order feature representation for a corresponding acoustic frame. The method also includes augmenting the sequence of acoustic frames and generating, as output form an unsupervised audio encoder, a predicted higher order feature representation for a corresponding augmented acoustic frame in the sequence of augmented acoustic frames. The method also includes determining an unsupervised loss term based on the target higher order feature representation and the predicted higher order feature representation and updating parameters of the speech recognition model based on the unsupervised loss term.

Semi-supervised training scheme for speech recognition

A transformer-transducer model for unifying streaming and non-streaming speech recognition includes an audio encoder, a label encoder, and a joint network. The audio encoder receives a sequence of acoustic frames, and generates, at each of a plurality of time steps, a higher order feature representation for a corresponding acoustic frame. The label encoder receives a sequence of non-blank symbols output by a final softmax layer, and generates, at each of the plurality of time steps, a dense representation. The joint network receives the higher order feature representation and the dense representation at each of the plurality of time steps, and generates a probability distribution over possible speech recognition hypothesis. The audio encoder of the model further includes a neural network having an initial stack of transformer layers trained with zero look ahead audio context, and a final stack of transformer layers trained with a variable look ahead audio context.

One model unifying streaming and non-streaming speech recognition

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

Systems and techniques are provided for sound model localization within an environment. Sound recordings of sounds in the environment may be received from devices in the environment. Preliminary labels for the sound recordings may be determined using pre-trained sound models. The preliminary labels may have associated probabilities. Sound clips with preliminary labels may be generated based on sound recordings that have preliminary labels whose probability is over a high-recall threshold for the pre-trained sound model that determined the preliminary label. The sound clips with preliminary labels may be sent to a user device. Labeled sound clips may be received from the user device. The labeled sound clips may be based on the sound clips with preliminary labels. Training data sets may be generated for the pre-trained sound models using the labeled sound clips. The pre-trained sound models may be trained using the training data sets to generate localized sound models.

Sound model localization within an environment

A streaming speech recognition model includes an audio encoder configured to receive a sequence of acoustic frames and generate a higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The streaming speech recognition model also includes a label encoder configured to receive a sequence of non-blank symbols output by a final softmax layer and generate a dense representation. The streaming speech recognition model also includes a joint network configured to receive the higher order feature representation generated by the audio encoder and the dense representation generated by the label encoder and generate a probability distribution over possible speech recognition hypotheses. Here, the streaming speech recognition model is trained using self-alignment to reduce prediction delay by encouraging an alignment path that is one frame left from a reference forced-alignment frame.

Reducing streaming ASR model delay with self alignment

Transformer transducer: one model unifying streaming and non-streaming speech recognition

Growing community of inventors

Mountain View, CA, United States of America

Qian Zhang