公开(公告)号：US20220067292A1

公开(公告)日：2022-03-03

申请号：US17007270

申请日：2020-08-31

Applicant: Google LLC

Inventor： Abhinav Rastogi ,  Mihir Sanjay Kale

IPC: G06F40/30 ,  G06F3/16

Abstract: Systems and methods for guided text generation in task-based dialogue. In some aspects of the technology, an automated assistant system is configured to receive a user request, call multiple APIs, generate dialogue acts based on data received from each API, replace any slot names in the dialogue acts with natural language descriptions of the slots, concatenate the modified dialogue acts, and pass the concatenated result to an NLG model for generation of a natural language response. In some aspects of the technology, the automated assistant may be configured to generate simple templated responses based on the data received from each API, concatenate the simple templated responses, and pass the concatenated sequence to an NLG model trained as a sequence-to-sequence transformer for generation of a final natural language response.

公开(公告)号：US11604929B2

公开(公告)日：2023-03-14

申请号：US17007270

申请日：2020-08-31

Applicant: Google LLC

Inventor： Abhinav Rastogi ,  Mihir Sanjay Kale

IPC: G06F17/00 ,  G06F40/30 ,  G06F3/16

Abstract: Systems and methods for guided text generation in task-based dialogue. In some aspects of the technology, an automated assistant system is configured to receive a user request, call multiple APIs, generate dialogue acts based on data received from each API, replace any slot names in the dialogue acts with natural language descriptions of the slots, concatenate the modified dialogue acts, and pass the concatenated result to an NLG model for generation of a natural language response. In some aspects of the technology, the automated assistant may be configured to generate simple templated responses based on the data received from each API, concatenate the simple templated responses, and pass the concatenated sequence to an NLG model trained as a sequence-to-sequence transformer for generation of a final natural language response.

公开(公告)号：US20250111161A1

公开(公告)日：2025-04-03

申请号：US18978233

申请日：2024-12-12

Applicant: Google LLC

Inventor： Abhinav Rastogi ,  Mihir Sanjay Kale

IPC: G06F40/30 ,  G06F3/16

Abstract: Systems and methods for guided text generation in task-based dialogue. In some aspects of the technology, an automated assistant system is configured to receive a user request, call multiple APIs, generate dialogue acts based on data received from each API, replace any slot names in the dialogue acts with natural language descriptions of the slots, concatenate the modified dialogue acts, and pass the concatenated result to an NLG model for generation of a natural language response. In some aspects of the technology, the automated assistant may be configured to generate simple templated responses based on the data received from each API, concatenate the simple templated responses, and pass the concatenated sequence to an NLG model trained as a sequence-to-sequence transformer for generation of a final natural language response.

公开(公告)号：US12197872B2

公开(公告)日：2025-01-14

申请号：US18167946

申请日：2023-02-13

Applicant: Google LLC

Inventor： Abhinav Rastogi ,  Mihir Sanjay Kale

IPC: G06F17/00 ,  G06F3/16 ,  G06F40/30

Abstract: Systems and methods for guided text generation in task-based dialogue. In some aspects of the technology, an automated assistant system is configured to receive a user request, call multiple APIs, generate dialogue acts based on data received from each API, replace any slot names in the dialogue acts with natural language descriptions of the slots, concatenate the modified dialogue acts, and pass the concatenated result to an NLG model for generation of a natural language response. In some aspects of the technology, the automated assistant may be configured to generate simple templated responses based on the data received from each API, concatenate the simple templated responses, and pass the concatenated sequence to an NLG model trained as a sequence-to-sequence transformer for generation of a final natural language response.

公开(公告)号：US20230186033A1

公开(公告)日：2023-06-15

申请号：US18167946

申请日：2023-02-13

Applicant: Google LLC

Inventor： Abhinav Rastogi ,  Mihir Sanjay Kale

IPC: G06F40/30 ,  G06F3/16

CPC classification number: G06F40/30 ,  G06F3/167

Abstract: Systems and methods for guided text generation in task-based dialogue. In some aspects of the technology, an automated assistant system is configured to receive a user request, call multiple APIs, generate dialogue acts based on data received from each API, replace any slot names in the dialogue acts with natural language descriptions of the slots, concatenate the modified dialogue acts, and pass the concatenated result to an NLG model for generation of a natural language response. In some aspects of the technology, the automated assistant may be configured to generate simple templated responses based on the data received from each API, concatenate the simple templated responses, and pass the concatenated sequence to an NLG model trained as a sequence-to-sequence transformer for generation of a final natural language response.