公开(公告)号：US20240211688A1

公开(公告)日：2024-06-27

申请号：US18545147

申请日：2023-12-19

Applicant: Google LLC

Inventor： Abhirut Gupta ,  Aravindan Raghuveer ,  Abhay Sharma ,  Nitin Raut ,  Manish Kumar

IPC: G06F40/284 ,  G06F40/232 ,  G06F40/242 ,  G10L13/08 ,  G10L15/02 ,  G10L15/06 ,  G10L15/187

CPC classification number: G06F40/284 ,  G06F40/232 ,  G06F40/242 ,  G10L13/08 ,  G10L15/187 ,  G10L2015/025 ,  G10L15/063

Abstract: Systems and methods for generating phonetic spelling variations of a given word based on locale-specific pronunciations. A phoneme-letter density model may be configured to identify a phoneme sequence corresponding to an input word, and to identify all character sequences that may correspond to an input phoneme sequence and their respective probabilities. The phoneme-phoneme error model may be configured to identify locale-specific alternative phoneme sequences that may correspond to a given phoneme sequence, and their respective probabilities. Using these two models, a processing system may be configured to generate, for a given input word, a list of alternative character sequences that may correspond to the input word based on locale-specific pronunciations, and/or a probability distribution representing how likely each alternative character sequence is to correspond to the input word.

公开(公告)号：US11893349B2

公开(公告)日：2024-02-06

申请号：US17716430

申请日：2022-04-08

Applicant: GOOGLE LLC

Inventor： Abhirut Gupta ,  Aravindan Raghuveer ,  Abhay Sharma ,  Nitin Raut ,  Manish Kumar

IPC: G06F40/232 ,  G10L15/187 ,  G06F40/284 ,  G06F40/242 ,  G10L13/08 ,  G10L15/02 ,  G10L15/06

CPC classification number: G06F40/284 ,  G06F40/232 ,  G06F40/242 ,  G10L13/08 ,  G10L15/187 ,  G10L15/063 ,  G10L2015/025

Abstract: Systems and methods for generating phonetic spelling variations of a given word based on locale-specific pronunciations. A phoneme-letter density model may be configured to identify a phoneme sequence corresponding to an input word, and to identify all character sequences that may correspond to an input phoneme sequence and their respective probabilities. The phoneme-phoneme error model may be configured to identify locale-specific alternative phoneme sequences that may correspond to a given phoneme sequence, and their respective probabilities. Using these two models, a processing system may be configured to generate, for a given input word, a list of alternative character sequences that may correspond to the input word based on locale-specific pronunciations, and/or a probability distribution representing how likely each alternative character sequence is to correspond to the input word.

公开(公告)号：US20220391588A1

公开(公告)日：2022-12-08

申请号：US17716430

申请日：2022-04-08

Applicant: GOOGLE LLC

Inventor： Abhirut Gupta ,  Aravindan Raghuveer ,  Abhay Sharma ,  Nitin Raut ,  Manish Kumar

IPC: G06F40/284 ,  G06F40/232 ,  G06F40/242

Abstract: Systems and methods for generating phonetic spelling variations of a given word based on locale-specific pronunciations. A phoneme-letter density model may be configured to identify a phoneme sequence corresponding to an input word, and to identify all character sequences that may correspond to an input phoneme sequence and their respective probabilities. The phoneme-phoneme error model may be configured to identify locale-specific alternative phoneme sequences that may correspond to a given phoneme sequence, and their respective probabilities. Using these two models, a processing system may be configured to generate, for a given input word, a list of alternative character sequences that may correspond to the input word based on locale-specific pronunciations, and/or a probability distribution representing how likely each alternative character sequence is to correspond to the input word.