Noise-Robust Speaker Recognition Combining Missing Data Techniques and Universal Background Modeling

T. May, S. van de Par, A. Kohlrausch

Research output: Contribution to journalJournal articleResearchpeer-review

1149 Downloads (Pure)

Abstract

Although the field of automatic speaker recognition (ASR) has been the subject of extensive research over the past decades, the lack of robustness against background noise has remained a major challenge. This paper describes a noise-robust speaker recognition system that combines missing data (MD) recognition with the adaptation of speaker models using a universal background model (UBM). For MD recognition, the identification of reliable and unreliable feature components is required. For this purpose, the signal-to-noise ratio (SNR) based mask estimation performance of various state-of-the art noise estimation techniques and noise reduction schemes is compared. Speaker recognition experiments show that the usage of a UBM in combination with missing data recognition yields substantial improvements in recognition performance, especially in the presence of highly non-stationary background noise at low SNRs.
Original languageEnglish
JournalI E E E Transactions on Audio, Speech and Language Processing
Volume20
Issue number1
Pages (from-to)108-121
ISSN1558-7916
DOIs
Publication statusPublished - 2012
Externally publishedYes

Fingerprint Dive into the research topics of 'Noise-Robust Speaker Recognition Combining Missing Data Techniques and Universal Background Modeling'. Together they form a unique fingerprint.

Cite this