Google Inc.

Ehsan Variani

Tara N Sainath

Arun Narayanan

Trevor D Strohman

Bhuvana Ramabhadran

Ron J Weiss

Ruoming Pang

Michiel A U Bacchiani

Samuel Bengio

Kevin William Wilson

Neeraj Gaur

Izhak Shafran

David Rybach

Tongzhou Chen

Christopher Walter George Thornton

Pedro Jose Moreno Mengibar

Rohit Prakash Prabhavalkar

Yanzhang He

Ananda Theertha Suresh

Mitchel Weintraub

Parisa Haghani

Rami Botros

Cyril Allauzen

Ignacio Lopez Moreno

Chung-Cheng Chiu

Xin Lei

Erik Mcdermott

Jiahui Yu

Weiran Wang

Mirkó Visontai

Yangzhang He

Charles Caleb Peyser

Ronny Huang

Sepand Mavandadi

A method includes generating, using an audio encoder, a higher-order feature representation for each acoustic frame in a sequence of acoustic frames; generating, using a decoder, based on the higher-order feature representation, a plurality of speech recognition hypotheses, each hypotheses corresponding to a candidate transcription of an utterance and having an associated first likelihood score; generating, using an external language model, for each speech recognition hypothesis, a second likelihood score; determining, using a learnable fusion module, for each speech recognition hypothesis, a set of fusion weights based on the higher-order feature representation and the speech recognition hypothesis; and generating, using the learnable fusion module, for each speech recognition hypothesis, a third likelihood score based on the first likelihood score, the second likelihood score, and the set of fusion weights, the audio encoder and decoder trained using minimum additive error rate training in the presence of the external language model.

Rare word recognition with LM-aware MWER training

A method includes receiving a sequence of acoustic frames extracted from audio data corresponding to an utterance. During a first pass, the method includes processing the sequence of acoustic frames to generate N candidate hypotheses for the utterance. During a second pass, and for each candidate hypothesis, the method includes: generating a respective un-normalized likelihood score; generating a respective external language model score; generating a standalone score that models prior statistics of the corresponding candidate hypothesis; and generating a respective overall score for the candidate hypothesis based on the un-normalized likelihood score, the external language model score, and the standalone score. The method also includes selecting the candidate hypothesis having the highest respective overall score from among the N candidate hypotheses as a final transcription of the utterance.

Multilingual re-scoring models for automatic speech recognition

An automated speech recognition (ASR) model includes a first encoder, a second encoder, and a decoder. The first encoder receives, as input, a sequence of acoustic frames, and generates, at each of a plurality of output steps, a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The second encoder receives, as input, the first higher order feature representation generated by the first encoder at each of the plurality of output steps, and generates, at each of the plurality of output steps, a second higher order feature representation for a corresponding first higher order feature frame. The decoder receives, as input, the second higher order feature representation generated by the second encoder at each of the plurality of output steps, and generates, at each of the plurality of time steps, a first probability distribution over possible speech recognition hypotheses.

Cascaded encoders for simplified streaming and non-streaming ASR

An ASR model includes a first encoder configured to receive a sequence of acoustic frames and generate a first higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The ASR model also includes a second encoder configured to receive the first higher order feature representation generated by the first encoder at each of the plurality of output steps and generate a second higher order feature representation for a corresponding first higher order feature frame. The ASR model also includes a decoder configured to receive the second higher order feature representation generated by the second encoder at each of the plurality of output steps and generate a first probability distribution over possible speech recognition hypothesis. The ASR model also includes a language model configured to receive the first probability distribution over possible speech hypothesis and generate a rescored probability distribution.

Efficient streaming non-recurrent on-device end-to-end model

Aspects of the present disclosure enable humanly-specified relationships to contribute to a mapping that enables compression of the output structure of a machine-learned model. An exponential model such as a maximum entropy model can leverage a machine-learned embedding and the mapping to produce a classification output. In such fashion, the feature discovery capabilities of machine-learned models (e.g., deep networks) can be synergistically combined with relationships developed based on human understanding of the structural nature of the problem to be solved, thereby enabling compression of model output structures without significant loss of accuracy. These compressed models provide improved applicability to 'on device' or other resource-constrained scenarios.

Exponential modeling with deep learning features

This specification describes computer-implemented methods and systems. One method includes receiving, by a neural network of a speech recognition system, first data representing a first raw audio signal and second data representing a second raw audio signal. The first raw audio signal and the second raw audio signal describe audio occurring at a same period of time. The method further includes generating, by a spatial filtering layer of the neural network, a spatial filtered output using the first data and the second data, and generating, by a spectral filtering layer of the neural network, a spectral filtered output using the spatial filtered output. Generating the spectral filtered output comprises processing frequency-domain data representing the spatial filtered output. The method still further includes processing, by one or more additional layers of the neural network, the spectral filtered output to predict sub-word units encoded in both the first raw audio signal and the second raw audio signal.

Enhanced multi-channel acoustic models

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for speech recognition using complex linear projection are disclosed. In one aspect, a method includes the actions of receiving audio data corresponding to an utterance. The method further includes generating frequency domain data using the audio data. The method further includes processing the frequency domain data using complex linear projection. The method further includes providing the processed frequency domain data to a neural network trained as an acoustic model. The method further includes generating a transcription for the utterance that is determined based at least on output that the neural network provides in response to receiving the processed frequency domain data.

Complex linear projection for acoustic modeling

Multichannel speech recognition using neural networks

Linear transformation for speech recognition modeling

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for inputting speech data that corresponds to a particular utterance to a neural network; determining an evaluation vector based on output at a hidden layer of the neural network; comparing the evaluation vector with a reference vector that corresponds to a past utterance of a particular speaker; and based on comparing the evaluation vector and the reference vector, determining whether the particular utterance was likely spoken by the particular speaker.

Speaker verification using neural networks

Growing community of inventors

Mountain View, CA, United States of America

Ehsan Variani