公开(公告)号：US20220229992A1

公开(公告)日：2022-07-21

申请号：US17589186

申请日：2022-01-31

Applicant: GOOGLE LLC

Inventor： Leonid Velikovich ,  Petar Aleksic ,  Pedro Moreno

IPC: G06F40/295 ,  G06F40/30 ,  G10L15/06 ,  G10L15/187 ,  G10L15/22

Abstract: Speech processing techniques are disclosed that enable determining a text representation of named entities in captured audio data. Various implementations include determining the location of a carrier phrase in a word lattice representation of the captured audio data, where the carrier phrase provides an indication of a named entity. Additional or alternative implementations include matching a candidate named entity with the portion of the word lattice, and augmenting the word lattice with the matched candidate named entity.

公开(公告)号：US11386889B2

公开(公告)日：2022-07-12

申请号：US16698280

申请日：2019-11-27

Applicant: Google LLC

Inventor： Petar Aleksic ,  Pedro J. Moreno Mengibar ,  Leonid Velikovich

IPC: G10L15/197 ,  G10L15/16 ,  G10L15/18 ,  G10L15/187

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing contextual grammar selection are disclosed. In one aspect, a method includes the actions of receiving audio data of an utterance. The actions include generating a word lattice that includes multiple candidate transcriptions of the utterance and that includes transcription confidence scores. The actions include determining a context of the computing device. The actions include based on the context of the computing device, identifying grammars that correspond to the multiple candidate transcriptions. The actions include determining, for each of the multiple candidate transcriptions, grammar confidence scores that reflect a likelihood that a respective grammar is a match for a respective candidate transcription. The actions include selecting, from among the candidate transcriptions, a candidate transcription. The actions further include providing, for output, the selected candidate transcription as a transcription of the utterance.

公开(公告)号：US20200175969A1

公开(公告)日：2020-06-04

申请号：US16698280

申请日：2019-11-27

Applicant: Google LLC

Inventor： Petar Aleksic ,  Pedro J. Moreno Mengibar ,  Leonid Velikovich

IPC: G10L15/197 ,  G10L15/16 ,  G10L15/187 ,  G10L15/18

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing contextual grammar selection are disclosed. In one aspect, a method includes the actions of receiving audio data of an utterance. The actions include generating a word lattice that includes multiple candidate transcriptions of the utterance and that includes transcription confidence scores. The actions include determining a context of the computing device. The actions include based on the context of the computing device, identifying grammars that correspond to the multiple candidate transcriptions. The actions include determining, for each of the multiple candidate transcriptions, grammar confidence scores that reflect a likelihood that a respective grammar is a match for a respective candidate transcription. The actions include selecting, from among the candidate transcriptions, a candidate transcription. The actions further include providing, for output, the selected candidate transcription as a transcription of the utterance.

公开(公告)号：US20220310082A1

公开(公告)日：2022-09-29

申请号：US17807208

申请日：2022-06-16

Applicant: Google LLC

Inventor： Petar Aleksic ,  Pedro J. Moreno Mengibar ,  Leonid Velikovich

IPC: G10L15/197 ,  G10L15/16 ,  G10L15/18 ,  G10L15/187

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing contextual grammar selection are disclosed. In one aspect, a method includes the actions of receiving audio data of an utterance. The actions include generating a word lattice that includes multiple candidate transcriptions of the utterance and that includes transcription confidence scores. The actions include determining a context of the computing device. The actions include based on the context of the computing device, identifying grammars that correspond to the multiple candidate transcriptions. The actions include determining, for each of the multiple candidate transcriptions, grammar confidence scores that reflect a likelihood that a respective grammar is a match for a respective candidate transcription. The actions include selecting, from among the candidate transcriptions, a candidate transcription. The actions further include providing, for output, the selected candidate transcription as a transcription of the utterance.

公开(公告)号：US10529322B2

公开(公告)日：2020-01-07

申请号：US15681801

申请日：2017-08-21

Applicant: Google LLC

Inventor： Petar Aleksic ,  Michael D. Riley ,  Pedro J. Moreno Mengibar ,  Leonid Velikovich

IPC: G10L15/04 ,  G10L15/18 ,  G10L15/22 ,  G10L15/197 ,  G06F17/27 ,  G10L15/14 ,  G10L15/193

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for tagging during speech recognition. A word lattice that indicates probabilities for sequences of words in an utterance is obtained. A conditional probability transducer that indicates a frequency that sequences of both the words and semantic tags for the words appear is obtained. The word lattice and the conditional probability transducer are composed to construct a word lattice that indicates probabilities for sequences of both the words in the utterance and the semantic tags for the words. The word lattice that indicates probabilities for sequences of both the words in the utterance and the semantic tags for the words is used to generate a transcription that includes the words in the utterance and the semantic tags for the words.

公开(公告)号：US20250061889A1

公开(公告)日：2025-02-20

申请号：US18934638

申请日：2024-11-01

Applicant: Google LLC

Inventor： Ágoston Weisz ,  Leonid Velikovich

IPC: G10L15/06 ,  G10L15/22 ,  G10L15/26 ,  G10L15/30

Abstract: A method includes receiving audio data corresponding to a query spoken and processing the audio data to generate multiple candidate hypotheses each represented by a respective sequence of hypothesized terms. For each candidate hypothesis, the method includes determining whether the sequence of hypothesized terms includes a source phrase from a list of phrase correction pairs. Each phrase correction pair includes a corresponding source phrase that was misrecognized and a corresponding target phrase replacing the source phrase. When the respective sequence of hypothesized terms includes the source phrase, the method includes generating a corresponding additional candidate hypothesis that replaces the source phrase. The method also includes ranking the multiple candidate hypotheses and each corresponding additional candidate hypothesis generated and generating a transcription of the query spoken by the user by selecting the highest ranking one of the multiple candidate hypotheses and each additional candidate hypothesis.

公开(公告)号：US20230186898A1

公开(公告)日：2023-06-15

申请号：US17644416

申请日：2021-12-15

Applicant: Google LLC

Inventor： Ágoston Weisz ,  Leonid Velikovich

IPC: G10L15/06 ,  G10L15/22 ,  G10L15/30 ,  G10L15/26

CPC classification number: G10L15/063 ,  G10L15/22 ,  G10L15/30 ,  G10L15/26

Abstract: A method includes receiving audio data corresponding to a query spoken and processing the audio data to generate multiple candidate hypotheses each represented by a respective sequence of hypothesized terms. For each candidate hypothesis, the method includes determining whether the sequence of hypothesized terms includes a source phrase from a list of phrase correction pairs. Each phrase correction pair includes a corresponding source phrase that was misrecognized and a corresponding target phrase replacing the source phrase. When the respective sequence of hypothesized terms includes the source phrase, the method includes generating a corresponding additional candidate hypothesis that replaces the source phrase. The method also includes ranking the multiple candidate hypotheses and each corresponding additional candidate hypothesis generated and generating a transcription of the query spoken by the user by selecting the highest ranking one of the multiple candidate hypotheses and each additional candidate hypothesis.

公开(公告)号：US20240428785A1

公开(公告)日：2024-12-26

申请号：US18824716

申请日：2024-09-04

Applicant: Google LLC

Inventor： Petar Aleksic ,  Pedro J. Moreno Mengibar ,  Leonid Velikovich

IPC: G10L15/197 ,  G10L15/16 ,  G10L15/18 ,  G10L15/187

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for implementing contextual grammar selection are disclosed. In one aspect, a method includes the actions of receiving audio data of an utterance. The actions include generating a word lattice that includes multiple candidate transcriptions of the utterance and that includes transcription confidence scores. The actions include determining a context of the computing device. The actions include based on the context of the computing device, identifying grammars that correspond to the multiple candidate transcriptions. The actions include determining, for each of the multiple candidate transcriptions, grammar confidence scores that reflect a likelihood that a respective grammar is a match for a respective candidate transcription. The actions include selecting, from among the candidate transcriptions, a candidate transcription. The actions further include providing, for output, the selected candidate transcription as a transcription of the utterance.

公开(公告)号：US12154549B2

公开(公告)日：2024-11-26

申请号：US17644416

申请日：2021-12-15

Applicant: Google LLC

Inventor： Ágoston Weisz ,  Leonid Velikovich

IPC: G10L15/08 ,  G10L15/06 ,  G10L15/183 ,  G10L15/22 ,  G10L15/26 ,  G10L15/30

Abstract: A method includes receiving audio data corresponding to a query spoken and processing the audio data to generate multiple candidate hypotheses each represented by a respective sequence of hypothesized terms. For each candidate hypothesis, the method includes determining whether the sequence of hypothesized terms includes a source phrase from a list of phrase correction pairs. Each phrase correction pair includes a corresponding source phrase that was misrecognized and a corresponding target phrase replacing the source phrase. When the respective sequence of hypothesized terms includes the source phrase, the method includes generating a corresponding additional candidate hypothesis that replaces the source phrase. The method also includes ranking the multiple candidate hypotheses and each corresponding additional candidate hypothesis generated and generating a transcription of the query spoken by the user by selecting the highest ranking one of the multiple candidate hypotheses and each additional candidate hypothesis.

公开(公告)号：US11797772B2

公开(公告)日：2023-10-24

申请号：US17589186

申请日：2022-01-31

Applicant: GOOGLE LLC

Inventor： Leonid Velikovich ,  Petar Aleksic ,  Pedro Moreno

IPC: G10L15/22 ,  G10L15/187 ,  G06F40/295 ,  G06F40/30 ,  G10L15/06

CPC classification number: G06F40/295 ,  G06F40/30 ,  G10L15/063 ,  G10L15/187 ,  G10L15/22

Abstract: Speech processing techniques are disclosed that enable determining a text representation of named entities in captured audio data. Various implementations include determining the location of a carrier phrase in a word lattice representation of the captured audio data, where the carrier phrase provides an indication of a named entity. Additional or alternative implementations include matching a candidate named entity with the portion of the word lattice, and augmenting the word lattice with the matched candidate named entity.