SPEAKER MODEL ADAPTATION BASED ON CONFIDENCE SCORE


Creative Commons License

Mengusoglu E.

TEHNICKI VJESNIK-TECHNICAL GAZETTE, vol.22, no.4, pp.873-878, 2015 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 22 Issue: 4
  • Publication Date: 2015
  • Doi Number: 10.17559/tv-20140120095957
  • Journal Name: TEHNICKI VJESNIK-TECHNICAL GAZETTE
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.873-878
  • Keywords: confidence measure, Fisher transformation, speaker adaptation, speaker verification, SPEECH RECOGNITION
  • TED University Affiliated: Yes

Abstract

Confidence measures are expected to give a measure of reliability on the result of a speech/speaker recognition system. Most commonly used confidence measures are based on posterior word or phoneme probabilities which can be obtained from the output of the recognizer. In this paper we introduced a linear interpretation of posterior probability based confidence measure by using inverse Fisher transformation. Speaker adaptation consists in updating model parameters of a speaker independent model to have a better representation of the current speaker. Confidence measures give more reliable selection criteria to select the utterances which best represent the speaker. A linear interpretation of confidence measure is very important to select the most representative data for adaptation.