Abstract
Bu çalışmada, arkaplan verisi ile gerçekleştirme
verisi arasında konuşulan dil anlamında bir uyumsuzluk olması durumunda Türkçe
konuşmalar için konuşmacı tanıma performansı incelenmiştir. Gauss karışım
modeli - genel arkaplan modeli sınıflandırıcısı ile mel-frekansı kepstral
katsayıları konuşmacılara özgü öznitelikler olarak seçilmiştir. 47 erkek ve 26
bayan konuşmacıdan oluşan Türkçe veritabanı ile yapılan deneylerde görülmüştür
ki arkaplan modelini eğitmek için kullanılan seslerin dili ile konuşmacı doğrulama
deneylerinde kullanılan dil farklı olduğunda konuşmacı doğrulama performansı
dramatik bir şekilde düşmektedir. Örneğin, erkek konuşmacılar için Türkçe ses
verileri ile arkaplan modeli eğitildiğinde %1.73 eşit hata oranı elde
edilirken, İngilizce sesler ile eğitildiğinde %12.34 eşit hata oranı elde
edilmiştir.
References
- Akbacak M. and Hansen, J. H. L, (2007) Language normalization for bilingual speaker recognition systems, IEEE International Conference on Acostics, Speech and Signal Processing, 257-260. doi:10.1109/ICASSP.2007.366898
- Benesty, J. Sondhi, M. M. and Huang, Y. A., (2007) Springer Handbook of Speech Processing, Springer-Verlag, New York.
- Bosaris Toolkit (2010). Access address: https://sites.google.com/site/bosaristoolkit/ (Accessed in 17.11.2016)
- Büyük, O., and Arslan, L. M., (2012a) Model selection and score normalization for text-dependent single utterance speaker verification, Turkish Journal of Electrical Engineering and Computer Science, 20(2), 1277-1295. doi:10.3906/elk-1103-35
- Büyük, O., and Arslan, L. M., (2012b) Combining log-spectral mean subtraction at different frequency resolutions for handset-channel compensation in single utterance speaker verification, IET Signal Processing, 6(9), 824-828. doi:10.1049/iet-spr.2011.0270
- Dempster, A. P., Laird, N. M., and Rubin, D. B., (1977) Maximum likelihood from incomplete data via EM algorithm, Journal of the Royal Statistical Society, 39(1), 1-38. doi:10.2307/2984875
- Hansen J. H.L and Hasan, T. (2015) Speaker recognition by machines and humans: A tutorial review, IEEE Signal Processing Magazine, 32(6), 74-99. doi:10.1109/MSP.2015.2462851
- Luengo, I., Navas, E., Sainz, I, Saratxaga, I., Sanchez, J., Odriozola, I and Hernaez, I. (2008) Text independent speaker identification in multilingual environments, LREC, 1814-1817.
- Ma, B. and Meng, H., (2004) English-Chinese bilingual text-independent speaker verification, IEEE International Conference on Acostics, Speech and Signal Processing, 293-296. doi: 10.1109/ICASSP.2004.1327105
- Ma, B., Meng, H. M., and Mak, M. -W., (2007) Effects of device mismatch, language mismatch and environmental mismatch on speaker verification, IEEE International Conference on Acostics, Speech and Signal Processing, 301-304. doi:10.1109/ICASSP.2007.366909
- Misra, A. and Hansen, J. H. L., (2014) Spoken language mismatch in speaker verification: An investigation with NIST-SRE and CRSS bi-ling corpora, Spoken Language Technology, 372-377. doi:10.1109/SLT.2014.7078603
- Reynolds, D. A., Rose, R. C., (1995) Robust text-independent speaker identification using Gaussian mixture speaker models, IEEE Transactions on Speech and Audio Processing, 3(1), 72-83. doi:10.1109/89.365379
- Reynolds, D. A., Quatieri, T. F., and Dunn, R. B., (2000) Speaker verification using adapted Gaussian mixture models, Digital Signal Processing, 10(1), 19-41. doi:10.1006/dspr.1999.0361
Abstract
In this paper, effect of language mismatch between background data
and evaluation data is analyzed for text-independent speaker recognition in
particular for Turkish spoken language. Gaussian mixture model with universal
background model (GMM-UBM) classifier is utilized using Mel-frequency cepstral
coefficients (MFCCs) as speaker-specific features. Experiments conducted on a Turkish
speech database consisting of 47 male and 26 female speakers reveals that Turkish
speaker recognition performance dramatically degrades in case of language
mismatch between UBM and the evaluation data. For example 1.73% and 12.34%
equal error rates (EERs) are obtained for male speakers when UBM is trained
using Turkish and English data, respectively.
References
- Akbacak M. and Hansen, J. H. L, (2007) Language normalization for bilingual speaker recognition systems, IEEE International Conference on Acostics, Speech and Signal Processing, 257-260. doi:10.1109/ICASSP.2007.366898
- Benesty, J. Sondhi, M. M. and Huang, Y. A., (2007) Springer Handbook of Speech Processing, Springer-Verlag, New York.
- Bosaris Toolkit (2010). Access address: https://sites.google.com/site/bosaristoolkit/ (Accessed in 17.11.2016)
- Büyük, O., and Arslan, L. M., (2012a) Model selection and score normalization for text-dependent single utterance speaker verification, Turkish Journal of Electrical Engineering and Computer Science, 20(2), 1277-1295. doi:10.3906/elk-1103-35
- Büyük, O., and Arslan, L. M., (2012b) Combining log-spectral mean subtraction at different frequency resolutions for handset-channel compensation in single utterance speaker verification, IET Signal Processing, 6(9), 824-828. doi:10.1049/iet-spr.2011.0270
- Dempster, A. P., Laird, N. M., and Rubin, D. B., (1977) Maximum likelihood from incomplete data via EM algorithm, Journal of the Royal Statistical Society, 39(1), 1-38. doi:10.2307/2984875
- Hansen J. H.L and Hasan, T. (2015) Speaker recognition by machines and humans: A tutorial review, IEEE Signal Processing Magazine, 32(6), 74-99. doi:10.1109/MSP.2015.2462851
- Luengo, I., Navas, E., Sainz, I, Saratxaga, I., Sanchez, J., Odriozola, I and Hernaez, I. (2008) Text independent speaker identification in multilingual environments, LREC, 1814-1817.
- Ma, B. and Meng, H., (2004) English-Chinese bilingual text-independent speaker verification, IEEE International Conference on Acostics, Speech and Signal Processing, 293-296. doi: 10.1109/ICASSP.2004.1327105
- Ma, B., Meng, H. M., and Mak, M. -W., (2007) Effects of device mismatch, language mismatch and environmental mismatch on speaker verification, IEEE International Conference on Acostics, Speech and Signal Processing, 301-304. doi:10.1109/ICASSP.2007.366909
- Misra, A. and Hansen, J. H. L., (2014) Spoken language mismatch in speaker verification: An investigation with NIST-SRE and CRSS bi-ling corpora, Spoken Language Technology, 372-377. doi:10.1109/SLT.2014.7078603
- Reynolds, D. A., Rose, R. C., (1995) Robust text-independent speaker identification using Gaussian mixture speaker models, IEEE Transactions on Speech and Audio Processing, 3(1), 72-83. doi:10.1109/89.365379
- Reynolds, D. A., Quatieri, T. F., and Dunn, R. B., (2000) Speaker verification using adapted Gaussian mixture models, Digital Signal Processing, 10(1), 19-41. doi:10.1006/dspr.1999.0361
Details
| Subjects | Engineering |
|---|---|
| Journal Section | Research Articles |
| Authors | |
| Publication Date | April 27, 2017 |
| Submission Date | December 5, 2016 |
| Acceptance Date | April 21, 2017 |
| Published in Issue | Year 2017 Volume: 22 Issue: 1 |
Cite
Announcements:
30.03.2021-Beginning with our April 2021 (26/1) issue, in accordance with the new criteria of TR-Dizin, the Declaration of Conflict of Interest and the Declaration of Author Contribution forms fulfilled and signed by all authors are required as well as the Copyright form during the initial submission of the manuscript. Furthermore two new sections, i.e. ‘Conflict of Interest’ and ‘Author Contribution’, should be added to the manuscript. Links of those forms that should be submitted with the initial manuscript can be found in our 'Author Guidelines' and 'Submission Procedure' pages. The manuscript template is also updated. For articles reviewed and accepted for publication in our 2021 and ongoing issues and for articles currently under review process, those forms should also be fulfilled, signed and uploaded to the system by authors.