Shinji Watanabe

US Patent:

20200058314, Feb 20, 2020

Filed:

Aug 16, 2018

Appl. No.:

15/998765

Inventors:

- Cambridge MA, US
Shinji Watanabe - Baltimore MD, US
John Hershey - Winchester MA, US
Gordon Wichem - Boston MA, US

International Classification:

G10L 19/032

Abstract:

Systems and methods for audio signal processing including an input interface to receive a noisy audio signal including a mixture of target audio signal and noise. An encoder to map each time-frequency bin of the noisy audio signal to one or more phase-related value from one or more phase quantization codebook of phase-related values indicative of the phase of the target signal. Calculate, for each time-frequency bin of the noisy audio signal, a magnitude ratio value indicative of a ratio of a magnitude of the target audio signal to a magnitude of the noisy audio signal. A filter to cancel the noise from the noisy audio signal based on the phase-related values and the magnitude ratio values to produce an enhanced audio signal. An output interface to output the enhanced audio signal.

US Patent:

20190318725, Oct 17, 2019

Filed:

Apr 13, 2018

Appl. No.:

15/952330

Inventors:

- Cambridge MA, US
Takaaki Hori - Lexington MA, US
Shane Settle - Moraga CA, US
Hiroshi Seki - Toyohashi, JP
Shinji Watanabe - Baltimore MD, US
John Hershey - Winchester MA, US

International Classification:

G10L 15/16
G10L 17/00
G10L 15/06
G10L 15/22

Abstract:

Systems and methods for a speech recognition system for recognizing speech including overlapping speech by multiple speakers. The system including a hardware processor. A computer storage memory to store data along with having computer-executable instructions stored thereon that, when executed by the processor is to implement a stored speech recognition network. An input interface to receive an acoustic signal, the received acoustic signal including a mixture of speech signals by multiple speakers, wherein the multiple speakers include target speakers. An encoder network and a decoder network of the stored speech recognition network are trained to transform the received acoustic signal into a text for each target speaker. Such that the encoder network outputs a set of recognition encodings, and the decoder network uses the set of recognition encodings to output the text for each target speaker. An output interface to transmit the text for each target speaker.

US Patent:

20190189115, Jun 20, 2019

Filed:

Dec 15, 2017

Appl. No.:

15/843055

Inventors:

- Cambridge MA, US
Shinji Watanabe - Baltimore MD, US
John Hershey - Winchester MA, US

International Classification:

G10L 15/16
G10L 15/22
G10L 15/14
G10L 15/02

Abstract:

A speech recognition system includes an input device to receive voice sounds, one or more processors, and one or more storage devices storing parameters and program modules including instructions which cause the one or more processors to perform operations. The operations include extracting an acoustic feature sequence from audio waveform data converted from the voice sounds, encoding the acoustic feature sequence into a hidden vector sequence using an encoder network having encoder network parameters, predicting first output label sequence probabilities by feeding the hidden vector sequence to a decoder network having decoder network parameters, predicting second output level sequence probabilities by a hybrid network using character-base language models (LMs) and word-level LMs; and searching, using a label sequence search module, for an output label sequence having a highest sequence probability by combining the first and second output label sequence probabilities provided from the decoder network and the hybrid network.

US Patent:

20180330718, Nov 15, 2018

Filed:

May 11, 2017

Appl. No.:

15/592527

Inventors:

- Cambridge MA, US
Shinji Watanabe - Arlington MA, US
John Hershey - Winchester MA, US

Assignee:

Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA

International Classification:

G10L 15/16
G10L 15/02
G10L 15/14
G10L 19/00
G06N 7/00
G06N 3/04
G06N 3/08

Abstract:

A speech recognition system includes an input device to receive voice sounds, one or more processors, and one or more storage devices storing parameters and program modules including instructions executable by the one or more processors. The instructions includes extracting an accoustic feature sequence from audio waveform data converted from the voice sounds encoding the acoustic feature sequence into a hidden vector sequence using an encoder network having encoder network parameters, predicting first output label sequence probabilities by feeding the hidden vector sequence to a decoder network having decoder network parameters, predicting second output label sequence probabilities by a connectionist temporal classification (CTC) module using CTC network parameters and the hidden vector sequence from the encoder network, and searching, using a label sequence search module, for an output label sequence having a highest sequence probability by combining the first and second output label sequence probabilities provided from the decoder network and the CTC module.

US Patent:

20180157743, Jun 7, 2018

Filed:

Dec 7, 2016

Appl. No.:

15/371513

Inventors:

- Cambridge MA, US
Chiori Hori - Lexington MA, US
Shinji Watanabe - Arlington MA, US
John Hershey - Winchester MA, US
Bret Harsham - Newton MA, US
Jonathan Le Roux - Arlington MA, US

Assignee:

Mitsubishi Electric Research Laboratories, Inc. - Cambridge MA

International Classification:

G06F 17/30
G06N 3/04
G06N 3/08

Abstract:

A method for performing multi-label classification includes extracting a feature vector from an input vector including input data by a feature extractor, determining, by a label predictor, a relevant vector including relevant labels having relevant scores based on the feature vector, updating a binary masking vector by masking pre-selected labels having been selected in previous label selections, applying the updated binary masking vector to the relevant vector such that the relevant label vector is updated to exclude the pre-selected labels from the relevant labels, and selecting a relevant label from the updated relevant label vector based on the relevant scores of the updated relevant label vector.

US Patent:

20170221474, Aug 3, 2017

Filed:

Feb 2, 2016

Appl. No.:

15/013239

Inventors:

- Cambridge MA, US
Chiori Hori - Lexington MA, US
Shinji Watanabe - Arlington MA, US
John Hershey - Winchester MA, US

International Classification:

G10L 15/06
G10L 15/16
G10L 15/02
G10L 15/08

Abstract:

A method and for training a language model to reduce recognition errors, wherein the language model is a recurrent neural network language model (RNNLM) by first acquiring training samples. An automatic speech recognition system (ASR) is appled to the training samples to produce recognized words and probabilites of the recognized words, and an N-best list is selected from the recognized words based on the probabilities. determining word erros using reference data for hypotheses in the N-best list. The hypotheses are rescored using the RNNLM. Then, we determine gradients for the hypotheses using the word errors and gradients for words in the hypotheses. Lastly, parameters of the RNNLM are updated using a sum of the gradients.

US Patent:

20170161256, Jun 8, 2017

Filed:

Dec 4, 2015

Appl. No.:

14/959132

Inventors:

- Cambridge MA, US
Takaaki Hori - Lexington MA, US
Shinji Watanabe - Arlington MA, US
John Hershey - Winchester MA, US

International Classification:

G06F 17/27
G06N 3/04
G06N 3/08
G10L 15/26

Abstract:

A method and system processes utterances that are acquired either from an automatic speech recognition (ASR) system or text. The utterances have associated identities of each party, such as role A utterances and role B utterances. The information corresponding to utterances, such as word sequence and identity, are converted to features. Each feature is received in an input layer of a neural network(NN). A dimensionality of each feature is reduced, in a projection layer of the NN, to produce a reduced dimensional feature. The reduced dimensional feature is processed to provide probabilities of labels for the utterances.

US Patent:

20170011741, Jan 12, 2017

Filed:

May 5, 2016

Appl. No.:

15/147382

Inventors:

- Cambridge MA, US
Jonathan Le Roux - Arlington MA, US
Shinji Watanabe - Arlington MA, US
Zhuo Chen - New York NY, US

International Classification:

G10L 15/16
G10L 15/06
G10L 15/02

Abstract:

A method distinguishes components of a signal by processing the signal to estimate a set of analysis features, wherein each analysis feature defines an element of the signal and has feature values that represent parts of the signal, processing the signal to estimate input features of the signal, and processing the input features using a deep neural network to assign an associative descriptor to each element of the signal, wherein a degree of similarity between the associative descriptors of different elements is related to a degree to which the parts of the signal represented by the elements belong to a single component of the signal. The similarities between associative descriptors are processed to estimate correspondences between the elements of the signal and the components in the signal. Then, the signal is processed using the correspondences to distinguish component parts of the signal.