Subband architecture for automatic speaker recognition.

DOI10.1016/S0165-1684(00)00033-5MaRDI QIDQ1575498zbMATH OpenFDO

Authors L. Besacier, Jean-François Bonastre

Publication date 21 August 2000

Published in Signal Processing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/S0165-1684(00)00033-5

zbMATH Keywords

CorrelationNoisy speechParallel modelRecombinationSpeaker identificationSubband architecture

Mathematics Subject Classification ID

Communication theory (94A05) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Review text

We present an original approach for automatic speaker identification especially applicable to environments which cause partial corruption of the frequency spectrum of the signal. The general principle is to split the whole frequency domain into several subbands on which statistical recognizers are independently applied and then recombined to yield a global score and a global recognition decision. The choice of the subband architecture and the recombination strategies are particularly discussed. This technique had been shown to be robust for speech recognition when a narrow band noise degradation occurs. We first objectively verify this robustness for the speaker identification task. We also study which information is really used to recognize speakers. For this, speaker identification experiments on independent subbands are conducted for 630 speakers of TIMIT and NTIMIT databases. The results show that the speaker specific information is not equally distributed among subbands. In particular, the low-frequency subbands (under 600 Hz) and the high-frequency subbands (over 3000 Hz) are more speaker-specific than middle-frequency ones. In addition, experiments on different subband system architectures show that the correlations between frequency channels are of prime importance for speaker recognition. Some of these correlations are lost when the frequency domain is divided into subbands. Consequently, we propose a particularly redundant parallel architecture for which most of the correlations are kept. The performances obtained with this new system, using linear recombination strategies, are equivalent to those of a conventional fullband recognizer on clean and telephone speech. Experiments on speech corrupted by unpredictable noise show a better adaptability of this approach in noisy environments, compared to a conventional device, especially when pruning of some recognizers is performed.

Recommendations

• scientific article; zbMATH DE number 1717341
• Improved data modeling for text-dependent speaker recognition using sub-band processing
• scientific article; zbMATH DE number 1717108
• Speaker-specific mapping for text-independent speaker recognition
• Line spectral frequency representation of subbands for speech recognition

Cited in

(12)

• Investigation on LP-residual representations for speaker identification
• Improved data modeling for text-dependent speaker recognition using sub-band processing
• scientific article; zbMATH DE number 1717108 (Why is no real title available?)
• scientific article; zbMATH DE number 1717341 (Why is no real title available?)
• Speaker recognition based on multi-subsystems likelihood scores fusion
• scientific article; zbMATH DE number 2036494 (Why is no real title available?)
• Speaker-specific mapping for text-independent speaker recognition
• scientific article; zbMATH DE number 1414402 (Why is no real title available?)
• Research on Robust Speaker Identification Based on Adaptive Histogram Equalization
• Robust speech recognition algorithm based on the Fletcher-Allen principle
• Line spectral frequency representation of subbands for speech recognition
• A comparative study of explicit and implicit modelling of subsegmental speaker-specific excitation source information

This page was built for publication: Subband architecture for automatic speaker recognition.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1575498)