Patents by Inventor Antoine Jean Bruguier
Antoine Jean Bruguier 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).
-
Patent number: 11948062Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a compressed recurrent neural network (RNN). One of the systems includes a compressed RNN, the compressed RNN comprising a plurality of recurrent layers, wherein each of the recurrent layers has a respective recurrent weight matrix and a respective inter-layer weight matrix, and wherein at least one of recurrent layers is compressed such that a respective recurrent weight matrix of the compressed layer is defined by a first compressed weight matrix and a projection matrix and a respective inter-layer weight matrix of the compressed layer is defined by a second compressed weight matrix and the projection matrix.Type: GrantFiled: December 4, 2020Date of Patent: April 2, 2024Assignee: Google LLCInventors: Ouais Alsharif, Rohit Prakash Prabhavalkar, Ian C. McGraw, Antoine Jean Bruguier
-
Patent number: 11942076Abstract: A method includes receiving audio data encoding an utterance spoken by a native speaker of a first language, and receiving a biasing term list including one or more terms in a second language different than the first language. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data to generate speech recognition scores for both wordpieces and corresponding phoneme sequences in the first language. The method also includes rescoring the speech recognition scores for the phoneme sequences based on the one or more terms in the biasing term list, and executing, using the speech recognition scores for the wordpieces and the rescored speech recognition scores for the phoneme sequences, a decoding graph to generate a transcription for the utterance.Type: GrantFiled: February 16, 2022Date of Patent: March 26, 2024Assignee: Google LLCInventors: Ke Hu, Golan Pundak, Rohit Prakash Prabhavalkar, Antoine Jean Bruguier, Tara N. Sainath
-
Publication number: 20240029718Abstract: A method includes processing, using a speech recognizer, a first portion of audio data to generate a first lattice, and generating a first partial transcription for an utterance based on the first lattice. The method includes processing, using the recognizer, a second portion of the data to generate, based on the first lattice, a second lattice representing a plurality of partial speech recognition hypotheses for the utterance and a plurality of corresponding speech recognition scores. For each particular partial speech recognition hypothesis, the method includes generating a corresponding re-ranked score based on the corresponding speech recognition score and whether the particular partial speech recognition hypothesis shares a prefix with the first partial transcription.Type: ApplicationFiled: July 13, 2023Publication date: January 25, 2024Applicant: Google LLCInventors: Antoine Jean Bruguier, David Qiu, Yangzhang He, Trevor Strohman
-
Publication number: 20230274736Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.Type: ApplicationFiled: May 4, 2023Publication date: August 31, 2023Applicant: Google LLCInventors: Rohit Prakash Prabhavalkar, Golan Pundak, Tara N. Sainath, Antoine Jean Bruguier
-
Patent number: 11664021Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.Type: GrantFiled: December 9, 2021Date of Patent: May 30, 2023Assignee: Google LLCInventors: Rohit Prakash Prabhavalkar, Golan Pundak, Tara N. Sainath, Antoine Jean Bruguier
-
Publication number: 20220238101Abstract: Two-pass automatic speech recognition (ASR) models can be used to perform streaming on-device ASR to generate a text representation of an utterance captured in audio data. Various implementations include a first-pass portion of the ASR model used to generate streaming candidate recognition(s) of an utterance captured in audio data. For example, the first-pass portion can include a recurrent neural network transformer (RNN-T) decoder. Various implementations include a second-pass portion of the ASR model used to revise the streaming candidate recognition(s) of the utterance and generate a text representation of the utterance. For example, the second-pass portion can include a listen attend spell (LAS) decoder. Various implementations include a shared encoder shared between the RNN-T decoder and the LAS decoder.Type: ApplicationFiled: December 3, 2020Publication date: July 28, 2022Inventors: Tara N. Sainath, Yanzhang He, Bo Li, Arun Narayanan, Ruoming Pang, Antoine Jean Bruguier, Shuo-yiin Chang, Wei Li
-
Publication number: 20220172706Abstract: A method includes receiving audio data encoding an utterance spoken by a native speaker of a first language, and receiving a biasing term list including one or more terms in a second language different than the first language. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data to generate speech recognition scores for both wordpieces and corresponding phoneme sequences in the first language. The method also includes rescoring the speech recognition scores for the phoneme sequences based on the one or more terms in the biasing term list, and executing, using the speech recognition scores for the wordpieces and the rescored speech recognition scores for the phoneme sequences, a decoding graph to generate a transcription for the utterance.Type: ApplicationFiled: February 16, 2022Publication date: June 2, 2022Applicant: Google LLCInventors: Ke Hu, Golan Pundak, Rohit Prakash Prabhavalkar, Antoine Jean Bruguier, Tara N. Sainath
-
Publication number: 20220101836Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.Type: ApplicationFiled: December 9, 2021Publication date: March 31, 2022Applicant: Google LLCInventors: Rohit Prakash Prabhavalkar, Golan Pundak, Tara N. Sainath, Antoine Jean Bruguier
-
Patent number: 11270687Abstract: A method includes receiving audio data encoding an utterance spoken by a native speaker of a first language, and receiving a biasing term list including one or more terms in a second language different than the first language. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data to generate speech recognition scores for both wordpieces and corresponding phoneme sequences in the first language. The method also includes rescoring the speech recognition scores for the phoneme sequences based on the one or more terms in the biasing term list, and executing, using the speech recognition scores for the wordpieces and the rescored speech recognition scores for the phoneme sequences, a decoding graph to generate a transcription for the utterance.Type: GrantFiled: April 28, 2020Date of Patent: March 8, 2022Assignee: Google LLCInventors: Ke Hu, Antoine Jean Bruguier, Tara N. Sainath, Rohit Prakash Prabhavalkar, Golan Pundak
-
Patent number: 11217231Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.Type: GrantFiled: April 30, 2020Date of Patent: January 4, 2022Assignee: Google LLCInventors: Rohit Prakash Prabhavalkar, Golan Pundak, Tara N. Sainath, Antoine Jean Bruguier
-
Publication number: 20210089916Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a compressed recurrent neural network (RNN). One of the systems includes a compressed RNN, the compressed RNN comprising a plurality of recurrent layers, wherein each of the recurrent layers has a respective recurrent weight matrix and a respective inter-layer weight matrix, and wherein at least one of recurrent layers is compressed such that a respective recurrent weight matrix of the compressed layer is defined by a first compressed weight matrix and a projection matrix and a respective inter-layer weight matrix of the compressed layer is defined by a second compressed weight matrix and the projection matrix.Type: ApplicationFiled: December 4, 2020Publication date: March 25, 2021Applicant: Google LLCInventors: Ouais Alsharif, Rohit Prakash Prabhavalkar, Ian C. McGraw, Antoine Jean Bruguier
-
Patent number: 10878319Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a compressed recurrent neural network (RNN). One of the systems includes a compressed RNN, the compressed RNN comprising a plurality of recurrent layers, wherein each of the recurrent layers has a respective recurrent weight matrix and a respective inter-layer weight matrix, and wherein at least one of recurrent layers is compressed such that a respective recurrent weight matrix of the compressed layer is defined by a first compressed weight matrix and a projection matrix and a respective inter-layer weight matrix of the compressed layer is defined by a second compressed weight matrix and the projection matrix.Type: GrantFiled: December 29, 2016Date of Patent: December 29, 2020Assignee: Google LLCInventors: Ouais Alsharif, Rohit Prakash Prabhavalkar, Ian C. McGraw, Antoine Jean Bruguier
-
Publication number: 20200402501Abstract: A method of biasing speech recognition includes receiving audio data encoding an utterance and obtaining a set of one or more biasing phrases corresponding to a context of the utterance. Each biasing phrase in the set of one or more biasing phrases includes one or more words. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data and grapheme and phoneme data derived from the set of one or more biasing phrases to generate an output of the speech recognition model. The method also includes determining a transcription for the utterance based on the output of the speech recognition model.Type: ApplicationFiled: April 30, 2020Publication date: December 24, 2020Applicant: Google LLCInventors: Rohit Prakash Prabhavalkar, Golan Pundak, Tara N. Sainath, Antoine Jean Bruguier
-
Publication number: 20200349923Abstract: A method includes receiving audio data encoding an utterance spoken by a native speaker of a first language, and receiving a biasing term list including one or more terms in a second language different than the first language. The method also includes processing, using a speech recognition model, acoustic features derived from the audio data to generate speech recognition scores for both wordpieces and corresponding phoneme sequences in the first language. The method also includes rescoring the speech recognition scores for the phoneme sequences based on the one or more terms in the biasing term list, and executing, using the speech recognition scores for the wordpieces and the rescored speech recognition scores for the phoneme sequences, a decoding graph to generate a transcription for the utterance.Type: ApplicationFiled: April 28, 2020Publication date: November 5, 2020Applicant: Google LLCInventors: Ke Hu, Antoine Jean Bruguier, Tara N. Sainath, Rohit Prakash Prabhavalkar, Golan Pundak
-
Patent number: 10484319Abstract: Methods and apparatus are disclosed for resolving multiple interpretations of an ambiguous temporal term of a resource to a subset of the multiple interpretations. In some implementations, a group of one or more messages is identified, an ambiguous temporal term of the messages determined, additional content of the messages determined, and multiple interpretations of the ambiguous temporal term resolved to a subset based on the additional content.Type: GrantFiled: March 21, 2019Date of Patent: November 19, 2019Assignee: GOOGLE LLCInventors: Bryan Christopher Horling, Ashutosh Shukla, Antoine Jean Bruguier
-
Patent number: 10277543Abstract: Methods and apparatus are disclosed for resolving multiple interpretations of an ambiguous temporal term of a resource to a subset of the multiple interpretations. In some implementations, a group of one or more messages is identified, an ambiguous temporal term of the messages determined, additional content of the messages determined, and multiple interpretations of the ambiguous temporal term resolved to a subset based on the additional content.Type: GrantFiled: June 26, 2014Date of Patent: April 30, 2019Assignee: GOOGLE LLCInventors: Bryan Christopher Horling, Ashutosh Shukla, Antoine Jean Bruguier
-
Publication number: 20190068532Abstract: Methods and apparatus are disclosed for resolving multiple interpretations of an ambiguous temporal term of a resource to a subset of the multiple interpretations. In some implementations, a group of one or more messages is identified, an ambiguous temporal term of the messages determined, additional content of the messages determined, and multiple interpretations of the ambiguous temporal term resolved to a subset based on the additional content.Type: ApplicationFiled: June 26, 2014Publication date: February 28, 2019Inventors: Bryan Christopher Horling, Ashutosh Shukla, Antoine Jean Bruguier
-
Patent number: 10152965Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage medium, for implementing a pronunciation dictionary that stores entity name pronunciations. In one aspect, a method includes actions of receiving audio data corresponding to an utterance that includes a command and an entity name. Additional actions may include generating, by an automated speech recognizer, an initial transcription for a portion of the audio data that is associated with the entity name, receiving a corrected transcription for the portion of the utterance that is associated with the entity name, obtaining a phonetic pronunciation that is associated with the portion of the audio data that is associated with the entity name, updating a pronunciation dictionary to associate the phonetic pronunciation with the entity name, receiving a subsequent utterance that includes the entity name, and transcribing the subsequent utterance based at least in part on the updated pronunciation dictionary.Type: GrantFiled: February 3, 2016Date of Patent: December 11, 2018Assignee: Google LLCInventors: Antoine Jean Bruguier, Fuchun Peng, Francoise Beaufays
-
Publication number: 20170221475Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage medium, for implementing a pronunciation dictionary that stores entity name pronunciations. In one aspect, a method includes actions of receiving audio data corresponding to an utterance that includes a command and an entity name. Additional actions may include generating, by an automated speech recognizer, an initial transcription for a portion of the audio data that is associated with the entity name, receiving a corrected transcription for the portion of the utterance that is associated with the entity name, obtaining a phonetic pronunciation that is associated with the portion of the audio data that is associated with the entity name, updating a pronunciation dictionary to associate the phonetic pronunciation with the entity name, receiving a subsequent utterance that includes the entity name, and transcribing the subsequent utterance based at least in part on the updated pronunciation dictionary.Type: ApplicationFiled: February 3, 2016Publication date: August 3, 2017Inventors: Antoine Jean Bruguier, Fuchun Peng, Francoise Beaufays
-
Publication number: 20170220925Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for implementing a compressed recurrent neural network (RNN). One of the systems includes a compressed RNN, the compressed RNN comprising a plurality of recurrent layers, wherein each of the recurrent layers has a respective recurrent weight matrix and a respective inter-layer weight matrix, and wherein at least one of recurrent layers is compressed such that a respective recurrent weight matrix of the compressed layer is defined by a first compressed weight matrix and a projection matrix and a respective inter-layer weight matrix of the compressed layer is defined by a second compressed weight matrix and the projection matrix.Type: ApplicationFiled: December 29, 2016Publication date: August 3, 2017Inventors: Ouais Alsharif, Rohit Prakash Prabhavalkar, Ian C. McGraw, Antoine Jean Bruguier