Abstract
The paper presents results of an evaluation of covariance matrix and i-vector based speaker identification methods on Serbian S70W100s120 database. Open set speaker identification evaluation scheme was adopted. The number of target speakers and the number of impostors were 20 and 60 respectively. Additional utterances from 41 speakers were used for training. Amount of data for modeling a target speaker was limited to about 4 s of speech. In this study, the i-vector base approach showed significantly better performance (equal error rate EER ~5%) than the covariance matrix based approach (EER ~16%). This small EER for the i-vector based approach was obtained after substantial reduction of the number of the parameters in universal background model, i-vector transformation matrix and Gaussian probabilistic linear discriminant analysis that is typically reported in the papers. Additionally, these experiments showed that cepstral mean and variance normalization can deteriorate EER in case of a single channel.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Hennerbert, J.: Speaker recognition, overview. In: Encyclopedia of Biometrics. Springer Science + Business Media, New York (2009)
Gonzalez-Rodriguez, J.: Evaluating automatic speaker recognition systems: an overview of the NIST speaker recognition evaluations (1996–2014). Loquens 1(1), e007 (2014)
Kohler, T.: The 2010 NIST Speaker Recognition Evaluation. http://archive.signalprocessingsociety.org/technical-committees/list/sl-tc/spl-nl/2010-07/NIST-SRE/. Accessed Mar 2017
McLaren, M., Ferrer, L., Castán, D., Lawson, A.: The 2016 speakers in the wild speaker recognition evaluation. In: INTERSPEECH 2016, San Francisco, CA, USA, pp. 823–827 (2016)
Matejka, P., Glembek, O., Castalado, F., Alam, M.J., Plchot, O., Kenny, P., Burget, L., Černocky, J.: Full-covariance UBM and heavy-tailed PLDA in i-vector speaker verification. In: ICASSP 2011, Prague, Czech Republic, pp. 4828–4831 (2011)
Jokić, I., Delić, V., Jokić, S., Perić, Z.: Automatic speaker recognition dependency on both the shape of auditory critical bands and speaker discriminative MFCCs. Adv. Electr. Comput. Eng. 15(4), 25–32 (2015)
Novotny, O., Matejka, P., Plchot, O., Glembek, O., Burget, L., Černocky, J.: Analysis of speaker recognition systems in realistic scenarios of the SITW 2016 challenge. In: INTERSPEECH 2016, San Francisco, CA, USA, pp. 828–832 (2016)
Sadjadi, S., Ganapathy, S., Pelecanos, J.: The IBM speaker recognition system: recent advances and error analysis. In: INTERSPEECH 2016, San Francisco, CA, USA, pp. 3633–3637 (2016)
Hasan, T., Liu, G., Sadjadi, S.O., Shokouhi, N., Boril, H., Ziaei, A., Misra, A., Godin, K.W., Hansen, J.: UTD-CRSS systems for 2012 NIST speaker recognition evaluation. In: ICASSP 2013, Vancouver, BC, Canada, pp. 6783–6787 (2013)
Garcia-Romero, D., Espy-Wilson, C: Analysis of i-vector length normalization in speaker recognition systems. In: INTERSPEECH 2011, Florence, Italy, pp. 249–252 (2011)
Wildermoth, B.: Text-Independent Speaker Recognition Using Source Based Features. Master thesis, Griffith University, Australia (2001)
Gelembek, O., Burget, L., Matejka, P., Karafiat, M., Kenny, P.: Simplification and optimization of i-vector extraction. In: ICASSP 2011, Prague, Czech Republic, pp. 4516–4519 (2011)
Kenny, P.: Joint factor analysis of speaker and session variability: Theory and algorithms. Technical report CRIM-06/08-13, CRIM, Montreal (2005)
Sadjadi, S., Slaney, M., Heck, L.: MSR Identity Toolbox: A MATLAB Toolbox for Speaker Recognition Research. Technical report, Microsoft Research, Conversational Systems Research Center (2013)
Brookes, M.: VOICEBOX. http://www.ee.ic.ac.uk/hp/staff/dmb/voicebox/voicebox.html
Delić, V., Sečujski, M., Jakovljević, N., Pekar, D., Mišković, D., Popović, B., Ostrogonac, S., Bojanić, M., Knežević, D.: Speech and language resources within speech recognition and synthesis systems for Serbian and Kindred South Slavic Languages. In: Železný, M., Habernal, I., Ronzhin, A. (eds.) SPECOM 2013. LNCS, vol. 8113, pp. 319–326. Springer, Cham (2013). doi:10.1007/978-3-319-01931-4_42
Acknowledgments
This research work has been supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, and it has been realized as a part of the research project TR 32035 and EUREKA project DANSPLAT (project ID 9944).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Jakovljević, N., Jokić, I., Jošić, S., Delić, V. (2017). A Comparison of Covariance Matrix and i-vector Based Speaker Recognition. In: Karpov, A., Potapova, R., Mporas, I. (eds) Speech and Computer. SPECOM 2017. Lecture Notes in Computer Science(), vol 10458. Springer, Cham. https://doi.org/10.1007/978-3-319-66429-3_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-66429-3_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-66428-6
Online ISBN: 978-3-319-66429-3
eBook Packages: Computer ScienceComputer Science (R0)