Speaker Detection Using Phoneme Specific Hidden Markov Models

  • Edvin Pakoci
  • Nikša Jakovljević
  • Branislav Popović
  • Dragiša Mišković
  • Darko Pekar
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8773)


The paper presents a speaker detection system based on phoneme specific hidden Markov model in combination with Gaussian mixture model. Our motivation stems from the fact that the phoneme specific HMM system can model temporal variations and provides possibility to ponder the scores of specific phonemes as well as efficient pruning. The performance of the system has been evaluated on speech database which contains utterances in Serbian from 250 speakers (10 of them being the target speakers). The proposed model is compared to a system based on Gaussian mixture model - universal background model, and showed a significant improvement in detection performance.


Speaker detection Hidden Markov models Gaussian mixture models 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Beigi, H.: Fundamentals of Speaker Recognition. Springer (2011)Google Scholar
  2. 2.
    Auckenthaler, R., Parris, E., Carey, M.: Improving a GMM speaker verification system by phonetic weighting. In: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 1999), vol. 1, pp. 313–316. Phoenix, Arizona (1999)Google Scholar
  3. 3.
    Kajarekar, S., Hermansky, H.: Speaker verification based on broad phonetic categories. In: A Speaker Odyssey - The Speaker Recognition Workshop (2001)Google Scholar
  4. 4.
    Hansen, E., Slyh, R., Anderson, T.: Speaker recognition using phoneme-specific GMMs. In: ODYSSEY 2004-The Speaker and Language Recognition Workshop, pp. 179–184 (2004)Google Scholar
  5. 5.
    Dunn, R., Reynolds, D., Quatieri, T.: Approaches to speaker detection and tracking in conversational speech. Digit. Signal Process. 10, 93–112 (2000)CrossRefGoogle Scholar
  6. 6.
    Kinnunen, T., Li, H.: An Overview of Text-Independent Speaker Recognition: From Features to Supervectors. Speech Commun 52, 12–40 (2010)CrossRefGoogle Scholar
  7. 7.
    Scheffer, N., Ferrer, L., Graciarena, M., Kajarekar, S., Shriberg, E., Stolcke, A.: The SRI NIST 2010 Speaker Recognition Evaluation System. In: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2011), pp. 5292–5295. Prague, Czech Republic (2011)CrossRefGoogle Scholar
  8. 8.
    Antal, M.: Phonetic Speaker Recognition. In: 7th International Conference COMMUNICATIONS, pp. 67–72 (2008)Google Scholar
  9. 9.
    Reynolds, D., Quatieri, T., Dunn, R.: Speaker Verification Using Adapted Gaussian Mixture Models. Digit. Signal Process. 10, 19–41 (2000)CrossRefGoogle Scholar
  10. 10.
    Delić, V., Sečujski, M., Jakovljević, N., Janev, M., Obradović, R., Pekar, D.: Speech Technologies for Serbian and Kindred South Slavic Languages. In: Advances in Speech Recognition, pp. 141–165 (2010)Google Scholar
  11. 11.
    Young, S.J., Evermann, G., Gales, M.J.F., Hain, T., Kershaw, D., Moore, G., Odell, J., Ollason, D., Povey, D., Valtchev, V., Woodland, P.C.: The HTK Book, version 3.4 (2006)Google Scholar
  12. 12.
    Gales, M., Young, S.: The Application of Hidden Markov Models in Speech Recognition. Foundations and Trends in Signal Processing 1(3), 195–304 (2007)CrossRefzbMATHGoogle Scholar
  13. 13.
    Jakovljević, N., Miškovic, D., Janev, M., Sečujski, M., Delić, V.: Comparison of Linear Discriminant Analysis Approaches in Automatic Speech Recognition. Elektronika Ir Elektrotechnika 19(7), 76–79 (2013)Google Scholar
  14. 14.
    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, Heidelberg (2013)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Edvin Pakoci
    • 1
  • Nikša Jakovljević
    • 1
  • Branislav Popović
    • 1
  • Dragiša Mišković
    • 1
  • Darko Pekar
    • 1
  1. 1.Faculty of Technical SciencesUniversity of Novi SadSerbia

Personalised recommendations