Google Inc.

Li Wan

Ignacio Lopez Moreno

Quan Wang

Alexander Houston Gruenstein

Jason William Pelecanos

Hakan Erdogan

Prashant Sridhar

Yang Yu

Hasim Sak

Yash Sheth

Alan Sean Papir

Techniques disclosed herein enable training and/or utilizing speaker dependent (SD) speech models which are personalizable to any user of a client device. Various implementations include personalizing a SD speech model for a target user by processing, using the SD speech model, a speaker embedding corresponding to the target user along with an instance of audio data. The SD speech model can be personalized for an additional target user by processing, using the SD speech model, an additional speaker embedding, corresponding to the additional target user, along with another instance of audio data. Additional or alternative implementations include training the SD speech model based on a speaker independent speech model using teacher student learning.

Speaker awareness using speaker dependent speech model(s)

Processing of acoustic features of audio data to generate one or more revised versions of the acoustic features, where each of the revised versions of the acoustic features isolates one or more utterances of a single respective human speaker. Various implementations generate the acoustic features by processing audio data using portion(s) of an automatic speech recognition system. Various implementations generate the revised acoustic features by processing the acoustic features using a mask generated by processing the acoustic features and a speaker embedding for the single human speaker using a trained voice filter model. Output generated over the trained voice filter model is processed using the automatic speech recognition system to generate a predicted text representation of the utterance(s) of the single human speaker without reconstructing the audio data.

Targeted voice separation by speaker for speech recognition

Methods and systems for training and/or using a language selection model for use in determining a particular language of a spoken utterance captured in audio data. Features of the audio data can be processed using the trained language selection model to generate a predicted probability for each of N different languages, and a particular language selected based on the generated probabilities. Speech recognition results for the particular language can be utilized responsive to selecting the particular language of the spoken utterance. Many implementations are directed to training the language selection model utilizing tuple losses in lieu of traditional cross-entropy losses. Training the language selection model utilizing the tuple losses can result in more efficient training and/or can result in a more accurate and/or robust model&#x2014;thereby mitigating erroneous language selections for spoken utterances.

Training and/or using a language selection model for automatically determining language for speech recognition of spoken utterance

Methods, systems, apparatus, including computer programs encoded on computer storage medium, to facilitate language independent-speaker verification. In one aspect, a method includes actions of receiving, by a user device, audio data representing an utterance of a user. Other actions may include providing, to a neural network stored on the user device, input data derived from the audio data and a language identifier. The neural network may be trained using speech data representing speech in different languages or dialects. The method may include additional actions of generating, based on output of the neural network, a speaker representation and determining, based on the speaker representation and a second representation, that the utterance is an utterance of the user. The method may provide the user with access to the user device based on determining that the utterance is an utterance of the user.

Speaker verification

Techniques are described for training and/or utilizing an end-to-end speaker diarization model. In various implementations, the model is a recurrent neural network (RNN) model, such as an RNN model that includes at least one memory layer, such as a long short-term memory (LSTM) layer. Audio features of audio data can be applied as input to an end-to-end speaker diarization model trained according to implementations disclosed herein, and the model utilized to process the audio features to generate, as direct output over the model, speaker diarization results. Further, the end-to-end speaker diarization model can be a sequence-to-sequence model, where the sequence can have variable length. Accordingly, the model can be utilized to generate speaker diarization results for any of various length audio segments.

Speaker diartzation using an end-to-end model

Speaker verification across locations, languages, and/or dialects

Improving speaker verification across locations, languages, and/or dialects

Systems, methods, devices, and other techniques for training and using a speaker verification neural network. A computing device may receive data that characterizes a first utterance. The computing device provides the data that characterizes the utterance to a speaker verification neural network. Subsequently, the computing device obtains, from the speaker verification neural network, a speaker representation that indicates speaking characteristics of a speaker of the first utterance. The computing device determines whether the first utterance is classified as an utterance of a registered user of the computing device. In response to determining that the first utterance is classified as an utterance of the registered user of the computing device, the device may perform an action for the registered user of the computing device.

Growing community of inventors

New York, NY, United States of America

Li Wan