Patent · US Active

Language agnostic multilingual end-to-end streaming on-device ASR system

US12183322B2 · kind B2 · utility

0Cited by
1References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateSep 22, 2022
Grant dateDec 31, 2024
Priority date
Expiry dateJun 27, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG10L2015/226
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method includes receiving a sequence of acoustic frames characterizing one or more utterances as input to a multilingual automated speech recognition (ASR) model. The method also includes generating a higher order feature representation for a corresponding acoustic frame. The method also includes generating a hidden representation based on a sequence of non-blank symbols output by a final softmax layer. The method also includes generating a probability distribution over possible speech recognition hypotheses based on the hidden representation generated by the prediction network at each of the plurality of output steps and the higher order feature representation generated by the encoder at each of the plurality of output steps. The method also includes predicting an end of utterance (EOU) token at an end of each utterance. The method also includes classifying each acoustic frame as either speech, initial silence, intermediate silence, or final silence.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.