Speaker indexing and adaptation using speaker clustering based on statistical model selection

• Published in: 2004 IEEE International Conference on Acoustics, Speech and Signal Processing Vol. 1; pp. I - 353
• Main Authors: Nishida, M., Kawahara, T.
• Format: Conference Proceeding
• Language: English; Japanese
• Published: Piscataway, N.J IEEE 28.09.2004
• Subjects: Acoustic testing; Applied sciences; Automatic speech recognition; Bayesian methods; Exact sciences and technology; Gaussian distribution; Indexing; Informatics; Information, signal and communications theory; Loudspeakers; Robustness; Signal and communications theory; Signal processing; Signal representation. Spectral analysis; Signal, noise; Speech processing; Speech recognition; Telecommunications and information theory; Voice mail; Performance evaluation; Automatic classification; Archive; Model selection; Mixture theory; Speaker adaptation; Unsupervised classification; Signal classification; Gaussian process; Statistical model; Automatic indexing; Speech recognition; Automatic recognition; Speech processing
• ISBN: 9780780384842; 0780384849
• ISSN: 1520-6149
• DOI: 10.1109/ICASSP.2004.1325995

The paper addresses unsupervised speaker indexing and automatic speech recognition of discussions. In speaker indexing, there are two cases, where the number of speakers is unknown beforehand and where the number is known. When the specified number is unknown, it is difficult to apply to various data because it needs to determine several parameters like threshold. In addition, serious problems arise in applying a uniform model because variations in the utterance durations of speakers are large. We thus propose a method which can robustly perform speaker indexing for the two cases using a flexible framework in which an optimal speaker model (GMM or VQ) is selected based on the BIC (Bayesian information criterion). Moreover, we propose a combination method of speaker adaptation based on speaker selection and the indexing method. For real discussion archives, we demonstrated that indexing performance is higher than that of conventional methods for the two cases and speech recognition performance was improved by the combination method.