Patents by Inventor Tara Sainath
Tara Sainath has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230352006Abstract: A RNN-T model includes a prediction network configured to, at each of a plurality of times steps subsequent to an initial time step, receive a sequence of non-blank symbols. For each non-blank symbol the prediction network is also configured to generate, using a shared embedding matrix, an embedding of the corresponding non-blank symbol, assign a respective position vector to the corresponding non-blank symbol, and weight the embedding proportional to a similarity between the embedding and the respective position vector. The prediction network is also configured to generate a single embedding vector at the corresponding time step. The RNN-T model also includes a joint network configured to, at each of the plurality of time steps subsequent to the initial time step, receive the single embedding vector generated as output from the prediction network at the corresponding time step and generate a probability distribution over possible speech recognition hypotheses.Type: ApplicationFiled: July 6, 2023Publication date: November 2, 2023Applicant: Google LLCInventors: Rami Botros, Tara Sainath
-
Publication number: 20230343328Abstract: 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.Type: ApplicationFiled: June 16, 2023Publication date: October 26, 2023Applicant: Google LLCInventors: Tara Sainath, Arun Narayanan, Rami Botros, Yanzhang He, Ehsan Variani, Cyril Allauzen, David Rybach, Ruoming Pang, Trevor Strohman
-
Patent number: 11727920Abstract: A RNN-T model includes a prediction network configured to, at each of a plurality of times steps subsequent to an initial time step, receive a sequence of non-blank symbols. For each non-blank symbol the prediction network is also configured to generate, using a shared embedding matrix, an embedding of the corresponding non-blank symbol, assign a respective position vector to the corresponding non-blank symbol, and weight the embedding proportional to a similarity between the embedding and the respective position vector. The prediction network is also configured to generate a single embedding vector at the corresponding time step. The RNN-T model also includes a joint network configured to, at each of the plurality of time steps subsequent to the initial time step, receive the single embedding vector generated as output from the prediction network at the corresponding time step and generate a probability distribution over possible speech recognition hypotheses.Type: GrantFiled: May 26, 2021Date of Patent: August 15, 2023Assignee: Google LLCInventors: Rami Botros, Tara Sainath
-
Patent number: 11715458Abstract: 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.Type: GrantFiled: May 10, 2021Date of Patent: August 1, 2023Assignee: Google LLCInventors: Tara Sainath, Arun Narayanan, Rami Botros, Yanzhang He, Ehsan Variani, Cyril Allauzen, David Rybach, Ruoming Pang, Trevor Strohman
-
Patent number: 11610586Abstract: A method includes receiving a speech recognition result, and using a confidence estimation module (CEM), for each sub-word unit in a sequence of hypothesized sub-word units for the speech recognition result: obtaining a respective confidence embedding that represents a set of confidence features; generating, using a first attention mechanism, a confidence feature vector; generating, using a second attention mechanism, an acoustic context vector; and generating, as output from an output layer of the CEM, a respective confidence output score for each corresponding sub-word unit based on the confidence feature vector and the acoustic feature vector received as input by the output layer of the CEM. For each of the one or more words formed by the sequence of hypothesized sub-word units, the method also includes determining a respective word-level confidence score for the word. The method also includes determining an utterance-level confidence score by aggregating the word-level confidence scores.Type: GrantFiled: February 23, 2021Date of Patent: March 21, 2023Assignee: Google LLCInventors: David Qiu, Qiujia Li, Yanzhang He, Yu Zhang, Bo Li, Liangliang Cao, Rohit Prabhavalkar, Deepti Bhatia, Wei Li, Ke Hu, Tara Sainath, Ian Mcgraw
-
Publication number: 20220310071Abstract: A RNN-T model includes a prediction network configured to, at each of a plurality of times steps subsequent to an initial time step, receive a sequence of non-blank symbols. For each non-blank symbol the prediction network is also configured to generate, using a shared embedding matrix, an embedding of the corresponding non-blank symbol, assign a respective position vector to the corresponding non-blank symbol, and weight the embedding proportional to a similarity between the embedding and the respective position vector. The prediction network is also configured to generate a single embedding vector at the corresponding time step. The RNN-T model also includes a joint network configured to, at each of the plurality of time steps subsequent to the initial time step, receive the single embedding vector generated as output from the prediction network at the corresponding time step and generate a probability distribution over possible speech recognition hypotheses.Type: ApplicationFiled: May 26, 2021Publication date: September 29, 2022Applicant: Google LLCInventors: Rami Botros, Tara Sainath
-
Publication number: 20220310062Abstract: 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.Type: ApplicationFiled: May 10, 2021Publication date: September 29, 2022Applicant: Google LLCInventors: Tara Sainath, Arun Narayanan, Rami Botros, Yangzhang He, Ehsan Variani, Cyrill Allauzen, David Rybach, Ruorning Pang, Trevor Strohman
-
Publication number: 20220270597Abstract: A method includes receiving a speech recognition result, and using a confidence estimation module (CEM), for each sub-word unit in a sequence of hypothesized sub-word units for the speech recognition result: obtaining a respective confidence embedding that represents a set of confidence features; generating, using a first attention mechanism, a confidence feature vector; generating, using a second attention mechanism, an acoustic context vector; and generating, as output from an output layer of the CEM, a respective confidence output score for each corresponding sub-word unit based on the confidence feature vector and the acoustic feature vector received as input by the output layer of the CEM. For each of the one or more words formed by the sequence of hypothesized sub-word units, the method also includes determining a respective word-level confidence score for the word. The method also includes determining an utterance-level confidence score by aggregating the word-level confidence scores.Type: ApplicationFiled: February 23, 2021Publication date: August 25, 2022Applicant: Google LLCInventors: David Qiu, Qiujia Li, Yanzhang He, Yu Zhang, Bo Li, Liangliang Cao, Rohit Prabhavalkar, Deepti Bhatia, Wei Li, Ke Hu, Tara Sainath, Ian Mcgraw
-
Publication number: 20220122622Abstract: 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.Type: ApplicationFiled: April 21, 2021Publication date: April 21, 2022Applicant: Google LLCInventors: Arun Narayanan, Tara Sainath, Chung-Cheng Chiu, Ruoming Pang, Rohit Prabhavalkar, Jiahui Yu, Ehsan Variani, Trevor Strohman
-
Publication number: 20220122586Abstract: A computer-implemented method of training a streaming speech recognition model that includes receiving, as input to the streaming speech recognition model, a sequence of acoustic frames. The streaming speech recognition model is configured to learn an alignment probability between the sequence of acoustic frames and an output sequence of vocabulary tokens. The vocabulary tokens include a plurality of label tokens and a blank token. At each output step, the method includes determining a first probability of emitting one of the label tokens and determining a second probability of emitting the blank token. The method also includes generating the alignment probability at a sequence level based on the first probability and the second probability. The method also includes applying a tuning parameter to the alignment probability at the sequence level to maximize the first probability of emitting one of the label tokens.Type: ApplicationFiled: September 9, 2021Publication date: April 21, 2022Applicant: Google LLCInventors: Jiahui Yu, Chung-cheng Chiu, Bo Li, Shuo-yiin Chang, Tara Sainath, Wei Han, Anmol Gulati, Yanzhang He, Arun Narayanan, Yonghui Wu, Ruoming Pang
-
Patent number: 10490198Abstract: A sensor device may include a computing device in communication with multiple microphones. A neural network executing on the computing device may receive audio signals from each microphone. One microphone signal may serve as a reference signal. The neural network may extract differences in signal characteristics of the other microphone signals as compared to the reference signal. The neural network may combine these signal differences into a lossy compressed signal. The sensor device may transmit the lossy compressed signal and the lossless reference signal to a remote neural network executing in a cloud computing environment for decompression and sound recognition analysis.Type: GrantFiled: December 18, 2017Date of Patent: November 26, 2019Assignee: GOOGLE LLCInventors: Chanwoo Kim, Rajeev Conrad Nongpiur, Tara Sainath
-
Publication number: 20180108363Abstract: A sensor device may include a computing device in communication with multiple microphones. A neural network executing on the computing device may receive audio signals from each microphone. One microphone signal may serve as a reference signal. The neural network may extract differences in signal characteristics of the other microphone signals as compared to the reference signal. The neural network may combine these signal differences into a lossy compressed signal. The sensor device may transmit the lossy compressed signal and the lossless reference signal to a remote neural network executing in a cloud computing environment for decompression and sound recognition analysis.Type: ApplicationFiled: December 18, 2017Publication date: April 19, 2018Inventors: CHANWOO KIM, Rajeev Conrad Nongpiur, TARA SAINATH
-
Patent number: 9875747Abstract: A sensor device may include a computing device in communication with multiple microphones. A neural network executing on the computing device may receive audio signals from each microphone. One microphone signal may serve as a reference signal. The neural network may extract differences in signal characteristics of the other microphone signals as compared to the reference signal. The neural network may combine these signal differences into a lossy compressed signal. The sensor device may transmit the lossy compressed signal and the lossless reference signal to a remote neural network executing in a cloud computing environment for decompression and sound recognition analysis.Type: GrantFiled: July 15, 2016Date of Patent: January 23, 2018Assignee: GOOGLE LLCInventors: Chanwoo Kim, Rajeev Conrad Nongpiur, Tara Sainath
-
Publication number: 20180018990Abstract: A sensor device may include a computing device in communication with multiple microphones. A neural network executing on the computing device may receive audio signals from each microphone. One microphone signal may serve as a reference signal. The neural network may extract differences in signal characteristics of the other microphone signals as compared to the reference signal. The neural network may combine these signal differences into a lossy compressed signal. The sensor device may transmit the lossy compressed signal and the lossless reference signal to a remote neural network executing in a cloud computing environment for decompression and sound recognition analysis.Type: ApplicationFiled: July 15, 2016Publication date: January 18, 2018Inventors: CHANWOO KIM, Rajeev Conrad Nongpiur, TARA SAINATH
-
Publication number: 20140164299Abstract: Pretraining for a DBN initializes weights of the DBN (Deep Belief Network) using a hybrid pre-training methodology. Hybrid pre-training employs generative component that allows the hybrid PT method to have better performance in WER (Word Error Rate) compared to the discriminative PT method. Hybrid pre-training learns weights which are more closely linked to the final objective function, allowing for a much larger batch size compared to generative PT, which allows for improvements in speed; and a larger batch size allows for parallelization of the gradient computation, speeding up training further.Type: ApplicationFiled: December 6, 2012Publication date: June 12, 2014Inventors: Tara Sainath, Brian Kingsbury, Bhuvana Ramabhadran