An Information Theoretic Approach to Speaker Diarization of Meeting Data

• Published in: IEEE transactions on audio, speech, and language processing speech, and language processing, 2009-09, Vol.17 (7), p.1382-1393
• Main Authors: Vijayasenan, D., Valente, F., Bourlard, H.
• Format: Article
• Language: English
• Subjects: Acoustic distortion; Applied sciences; Automatic speech recognition; Error analysis; Exact sciences and technology; Hidden Markov models; Indexing; Information bottleneck (IB); Information theory; Information, signal and communications theory; Loudspeakers; meetings data; Mutual information; NIST; Signal and communications theory; Signal processing; Signal representation. Spectral analysis; Signal, noise; speaker diarization; Speech processing; Streaming media; Studies; Telecommunications and information theory; Vocabulary
• ISSN: 1558-7916; 2329-9290; 1558-7924; 2329-9304
• DOI: 10.1109/TASL.2009.2015698

A speaker diarization system based on an information theoretic framework is described. The problem is formulated according to the information bottleneck (IB) principle. Unlike other approaches where the distance between speaker segments is arbitrarily introduced, the IB method seeks the partition that maximizes the mutual information between observations and variables relevant for the problem while minimizing the distortion between observations. This solves the problem of choosing the distance between speech segments, which becomes the Jensen-Shannon divergence as it arises from the IB objective function optimization. We discuss issues related to speaker diarization using this information theoretic framework such as the criteria for inferring the number of speakers, the tradeoff between quality and compression achieved by the diarization system, and the algorithms for optimizing the objective function. Furthermore, we benchmark the proposed system against a state-of-the-art system on the NIST RT06 (rich transcription) data set for speaker diarization of meetings. The IB-based system achieves a diarization error rate of 23.2% compared to 23.6% for the baseline system. This approach being mainly based on nonparametric clustering, it runs significantly faster than the baseline HMM/GMM based system, resulting in faster-than-real-time diarization.