New Background Speaker Models and Experiments on the ANDOSL Speech Corpus

We present a new background speaker modeling method for speaker verification. This method does not require speech databases of other speakers to build background speaker models. Background speaker models can be built directly from the same claimed speaker’s database and have smaller numbers of Gaussian mixtures compared to the claimed speaker model. Experiments performed on the Australian ANDOSL speech database showed better results for speaker verification using the 32-mixture claimed speaker model and 16-mixture background model compared to current background model set methods using five closest and five same-group background models.

[1]  Thomas Hain,et al.  The 1997 HTK broadcast news transcription system , 1998 .

[2]  Chin-Hui Lee,et al.  Background model design for flexible and portable speaker verification systems , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[3]  J.B. Millar,et al.  The Australian National Database of Spoken Language , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[4]  Douglas A. Reynolds,et al.  Speaker Verification Using Adapted Gaussian Mixture Models , 2000, Digit. Signal Process..

[5]  Dat Tran.  Fuzzy Normalisation Methods for Pattern Verification , 2004, ICBA.

[6]  Chin-Hui Lee,et al.  A new approach to utterance verification based on neighborhood information in model space , 2003, IEEE Trans. Speech Audio Process..

[7]  Corneliu Burileanu,et al.  On Performance Improvement of a Speaker Verification System Using Vector Quantization, Cohorts and Hybrid Cohort-World Models , 2002, Int. J. Speech Technol..

[8]  Seiichi Nakagawa,et al.  Speaker verification using frame and utterance level likelihood normalization , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[9]  Lawrence G. Bahler,et al.  Speaker verification using randomized phrase prompting , 1991, Digit. Signal Process..

[10]  Chin-Hui Lee,et al.  Speaker verification using normalized log-likelihood score , 1996, IEEE Trans. Speech Audio Process..

[11]  Jun-ichi Takahashi,et al.  A new cohort normalization using local acoustic information for speaker verification , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[12]  Douglas A. Reynolds,et al.  Speaker identification and verification using Gaussian mixture speaker models , 1995, Speech Commun..

[13]  Sadaoki Furui,et al.  Concatenated phoneme models for text-variable speaker recognition , 1993, 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing.