Patent · US Active

ASR training and adaptation

US10984801B2 · kind B2 · utility

0Cited by

2References

9Claims

0Family size

Assignee

Telefonaktiebolaget LM Ericsson (Publ) · SE

Inventors

Volodya Grancharov · Solna, SE
Erlendur Karlsson · Uppsala, SE
Sigurdur Sverrisson · Oststraße, SE
Maxim Teslenko · Sollentuna, SE
Konstantinos Vandikas · Solna, SE
Aneta Vulgarakis Feljan · Stockholm, SE

Key dates

Filing date	May 8, 2017
Grant date	Apr 20, 2021
Priority date	—
Expiry date	May 8, 2037

Classification

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

Abstract

AM and LM parameters to be used for adapting an ASR model are derived for each audio segment of an audio stream comprising multiple audio programs. A set of identifiers, including a speaker identifier, a speaker domain identifier and a program domain identifier, is obtained for each audio segment. The set of identifiers are used to select most suitable AM and LM parameters for the particular audio segment. The embodiments enable provision of maximum constraints on the AMs and LMs and enable adaptation of the ASR model on the fly for audio streams of multiple audio programs, such as broadcast audio. This means that the embodiments enable selecting AM and LM parameters that are most suitable in terms of ASR performance for each audio segment.

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