公开(公告)号：US20230335117A1

公开(公告)日：2023-10-19

申请号：US18186872

申请日：2023-03-20

Applicant: Google LLC

Inventor： Shuo-yiin Chang ,  Guru Prakash Arumugam ,  Zelin Wu ,  Tara N. Sainath ,  Bo LI ,  Qiao Liang ,  Adam Stambler ,  Shyam Upadhyay ,  Manaal Faruqui ,  Trevor Strohman

IPC: G10L15/16 ,  G10L15/22 ,  G10L15/06

CPC classification number: G10L15/16 ,  G10L15/22 ,  G10L15/063 ,  G10L2015/223

Abstract: A method includes receiving, as input to a speech recognition model, audio data corresponding to a spoken utterance. The method also includes performing, using the speech recognition model, speech recognition on the audio data by, at each of a plurality of time steps, encoding, using an audio encoder, the audio data corresponding to the spoken utterance into a corresponding audio encoding, and decoding, using a speech recognition joint network, the corresponding audio encoding into a probability distribution over possible output labels. At each of the plurality of time steps, the method also includes determining, using an intended query (IQ) joint network configured to receive a label history representation associated with a sequence of non-blank symbols output by a final softmax layer, an intended query decision indicating whether or not the spoken utterance includes a query intended for a digital assistant.

公开(公告)号：US20230289538A1

公开(公告)日：2023-09-14

申请号：US17981016

申请日：2022-11-04

Applicant: Google LLC

Inventor： Rahul Goel ,  Shyam Upadhyay ,  Anmol Agarwal

IPC: G06F40/58 ,  G06F40/205 ,  G06F40/30

CPC classification number: G06F40/58 ,  G06F40/205 ,  G06F40/30

Abstract: Systems and methods for generating code-switched semantic parsing training data and training of semantic parsers. In some examples, a processing system may be configured to use a trained first language model to translate a first single-language text sequence and first parsing data into a second code-switched text sequence and associated second parsing data, and to generate a second training example based on the second code-switched text sequence and the second parsing data. In some examples, the processing system may be further configured to generate a training set from two or more of these second training examples, and to use the training set to train a semantic parser to semantically parse code-switched utterances.

公开(公告)号：US20250124316A1

公开(公告)日：2025-04-17

申请号：US18899988

申请日：2024-09-27

Applicant: GOOGLE LLC

Inventor： Srividya Pranavi Potharaju ,  Shyam Upadhyay ,  Aman Madaan ,  Ankit Anand ,  Manaal Faruqui

IPC: G06N7/01

Abstract: Various implementations are directed towards generating, based on processing language model (LM) input using a first LM, an initial response that is predicted to be responsive to natural language (NL) based input, where the LM input includes at least the NL based input. Additionally or alternatively, the system can determine whether to generate an additional response based on processing the LM input using a second LM, where determining whether to generate the additional response includes processing at least the LM input and initial response using at least one verifier to generate a verification score. In many implementations, the verification score can be processed using a meta-verifier to determine whether to render output based on the initial response or the additional response.