Patent · US Active

Confidence features for automated speech recognition arbitration

US10706852B2 · kind B2 · utility

2Cited by

19References

22Claims

0Family size

Assignee

MICROSOFT TECHNOLOGY LICENSING, LLC · US

Inventors

Kshitiz Kumar · Redmond, US
Hosam A. Khalil · Redmond, US
Yifan Gong · Sammamish, US
Ziad Al-Bawab · San Jose, US
Chaojun Liu · Redmond, US

Key dates

Filing date	Nov 13, 2015
Grant date	Jul 7, 2020
Priority date	—
Expiry date	Jun 10, 2037

Classification

Technology area (CPC G)Physics
CPC primaryG10L15/183
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

The described technology provides arbitration between speech recognition results generated by different automatic speech recognition (ASR) engines, such as ASR engines trained according to different language or acoustic models. The system includes an arbitrator that selects between a first speech recognition result representing an acoustic utterance as transcribed by a first ASR engine and a second speech recognition result representing the acoustic utterance as transcribed by a second ASR engine. This selection is based on a set of confidence features that is initially used by the first ASR engine or the second ASR engine to generate the first and second speech recognition results.

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