Abstract
This paper is concerned with the speaker diarization task in the specific context of the meeting room recordings. Firstly, different technical improvements of an E-HMM based system are proposed and evaluated in the framework of the NIST RT’06S evaluation campaign. Related experiments show an absolute gain of 6.4% overall speaker diarization error rate (DER) and 12.9% on the development and evaluation corpora respectively.
Secondly, this paper presents an original strategy to deal with the overlapping speech. Indeed, speech overlaps between speakers are largely involved in meetings due to the spontaneous nature of this kind of data and they are responsible for a decrease in performance of the speaker diarization system, if they are not dealt with. Experiments still conducted in the framework of the NIST RT’06S evaluation show the ability of the strategy in detecting overlapping speech (decrease of the missed speaker error rate), even if an overall gain in speaker diarization performance has not been achieved yet.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
AMI: Augmented Multi-party Interaction project, http://www.amiproject.org/
CHIL: Computers in the Human Interaction Loop project, http://chil.server.de/servlet/is/101/
NIST: Spring 2006 (RT 2006S) Rich Transcription meeting recognition ev- aluation plan (2006), http://www.nist.gov/speech/tests/rt/rt2006/spring/docs/rt06s-meeting-eval-plan-V2.pdf
Istrate, D., Fredouille, C., Meignier, S., Besacier, L., Bonastre, J.F.: NIST RT 2005S evaluation: pre-processing techniques and speaker diarization on multiple microphone meetings. In: Renals, S., Bengio, S. (eds.) MLMI 2005. LNCS, vol. 3869. Springer, Heidelberg (2006)
Meignier, S., Moraru, D., Fredouille, C., Bonastre, J.F., Besacier, L.: Step-by-step and integrated approaches in broadcast news speaker diarization. Special issue of Computer and Speech Language Journal 20(2-3) (2006)
Moraru, D., Meignier, S., Fredouille, C., Besacier, L., Bonastre, J.F.: The ELISA consortium approaches in broadcast news speaker segmentation during the NIST 2003 rich transcription evaluation. In: ICASSP 2004, Montreal, Canada (2004)
Zhu, X., Barras, C., Meignier, S., Gauvain, J.L.: Combining speaker identification and BIC for speaker diarization. In: EuroSpeech 2005, Lisboa, Portugal (2005)
Bonastre, J.F., Wils, F., Meignier, S.: ALIZE, a free toolkit for speaker recognition. In: ICASSP 2005, Philadelphia, USA (2005)
Meignier, S., Bonastre, J.F., Fredouille, C., Merlin, T.: Evolutive HMM for speaker tracking system. In: ICASSP 2000, Istanbul, Turkey (2000)
Anguerra, X., Wooters, C., Peskin, B., Aguilo, M.: Robust speaker segmentation for meetings: the icsi-sri spring 2005 diarization system. In: Rich Transcription 2005 Spring Meeting Recognition Evaluation, Edinburgh, Scotland (2005)
Ajmera, J., Wooters, C.: A robust speaker clustering algorithm. In: ASRU 2003, US Virgin Islands, USA (2003)
Reynolds, D.A., Quatieri, T.F., Dunn, R.B.: Speaker verification using adapted gaussian mixture models. Digital Signal Processing (DSP), a review journal - Special issue on NIST 1999 speaker recognition workshop 10, 19–41 (2000)
Pelecanos, J., Sridharan, S.: Feature warping for robust speaker verification. In: 2001: A Speaker Odyssey. The Speaker Recognition Workshop, Chania, Crete, pp. 213–218 (2001)
Reynolds, D.A.: Channel robust speaker verification via feature mapping. In: ICASSP 2003 Conference, Hong Kong, China (2003)
Bonastre, J.F., Fredouille, C., Scheffer, N.: LIA 2005 system description. In: NIST SRE 2005 Workshop: speaker recognition evaluation campaign, Montreal, Canada (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Fredouille, C., Senay, G. (2006). Technical Improvements of the E-HMM Based Speaker Diarization System for Meeting Records. In: Renals, S., Bengio, S., Fiscus, J.G. (eds) Machine Learning for Multimodal Interaction. MLMI 2006. Lecture Notes in Computer Science, vol 4299. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11965152_32
Download citation
DOI: https://doi.org/10.1007/11965152_32
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-69267-6
Online ISBN: 978-3-540-69268-3
eBook Packages: Computer ScienceComputer Science (R0)