Speaker Diarization for Multi-microphone Meetings Using Only Between-Channel Differences

  • Jose M. Pardo
  • Xavier Anguera
  • Chuck Wooters
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4299)


We present a method to extract speaker turn segmentation from multiple distant microphones (MDM) using only delay values found via a cross-correlation between the available channels. The method is robust against the number of speakers (which is unknown to the system), the number of channels, and the acoustics of the room. The delays between channels are processed and clustered to obtain a segmentation hypothesis. We have obtained a 31.2% diarization error rate (DER) for the NIST´s RT05s MDM conference room evaluation set. For a MDM subset of NIST´s RT04s development set, we have obtained 36.93% DER and 35.73% DER*. Comparing those results with the ones presented by Ellis and Liu [8], who also used between-channels differences for the same data, we have obtained 43% relative improvement in the error rate.


False Alarm Broadcast News Speaker Diarization Segmentation Hypothesis Reference Microphone 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Ferreiros, J., Ellis, D.: Using Acoustic Condition Clustering To Improve Acoustic Change Detection On Broadcast News. In: Proc. ICSLP 2000 (2000)Google Scholar
  2. 2.
    Ajmera, J., Wooters, C.: A Robust speaker clustering algorithm. IEEE ASRU 2003 (2003)Google Scholar
  3. 3.
    Anguera, X., Wooters, C., Pesking, B., Aguiló, M.: Robust Speaker Segmentation for Meetings: The ICSI-SRI Spring 2005 Diarization System. In: Proc NIST MLMI Meeting Recognition Workshop, Edinburgh (2005)Google Scholar
  4. 4.
    Wooters, C., Mirghafori, N., Stolcke, A., Pirinen, T., Bulyko, I., Gelbart, D., Graciarena, M., Otterson, S., Peskin, B., Ostendorf, M.: The 2004 ICSI-SRI-UW Meeting Recognition System. In: Bengio, S., Bourlard, H. (eds.) MLMI 2004. LNCS, vol. 3361, pp. 196–208. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  5. 5.
    Wooters, C., Fung, J., Pesking, B., Anguera, X.: Towards Robust Speaker Segmentation: The ICSI-SRI Fall 2004 Diarization System. In: NIST RT-04F Workshop (November 2004)Google Scholar
  6. 6.
    Stolcke, A., Anguera, X., Boakye, K., Cetin, O., Grezl, F., Janin, A., Mandal, A., Peskin, B., Wooters, C., Zheng, J.: Further Progress in Meeting Recognition: The ICSI-SRI Spring 2005 Speech-to-Text Evaluation System. In: Proceedings of NIST MLMI Meeting Recognition Workshop, Edinburgh (2005)Google Scholar
  7. 7.
    Janin, A., Ang, J., Bhagat, S., Dhillon, R., Edwards, J., Macias-Guarasa, J., Morgan, N., Peskin, B., Shriberg, E., Stolcke, A., Wooters, C., Wrede, B.: The ICSI Meeting Project: Resources and Research. In: NIST ICASSP 2004 Meeting Recognition Workshop, Montreal (2004)Google Scholar
  8. 8.
    Elis, D.P.W., Liu, J.C.: Speaker Turn Segmentation Based On Between-Channels Differences. In: Proc. ICASSP 2004 (2004)Google Scholar
  9. 9.
    Anguera, X., Wooters, C., Hernando, J.: Speaker Diarization For Multi-Party Meetings Using Acoustic Fusion. IEEE ASRU (2005)Google Scholar
  10. 10.
    NIST Spring 2005 (RT05S) Rich Transcription Meeting Recognition Evaluation Plan (2005),
  11. 11.
    Brandstein, M.S., Silverman, H.F.: A Robust Method For Speech Signal Time-Delay Estimation In Reverberant Rooms. In: ICASSP 1997, Munich (1997)Google Scholar
  12. 12.
    Chen, S.S., Gopalakrishnan, P.S.: Speaker Environment And Channel Change Detection And Clustering Via The Bayesian Information Criterion. In: Proceedings DARPA Broadcast News Transcription and Understanding Workshop, Virginia, USA (February 1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Jose M. Pardo
    • 1
    • 2
  • Xavier Anguera
    • 1
    • 3
  • Chuck Wooters
    • 1
  1. 1.International Computer Science InstituteBerkeleyUSA
  2. 2.Universidad Politécnica de MadridMadridSpain
  3. 3.Technical University of CataloniaBarcelonaSpain

Personalised recommendations