Sixth International Conference on Spoken Language Processing
(ICSLP 2000)

Beijing, China
October 16-20, 2000

Speaker Change Detection Using Minimum Message Length Criterion

Chaojun Liu (1), Yonghong Yan (2)

(1) Center for Spoken Language Understanding, Oregon Graduate Institute, USA
(2) Intel Corporation, USA

Speaker change detection or speaker-based segmentation is useful and important in many applications, such as transcribing broadcast news or telephone conversations. It usually serves as a preliminary step prior to speech/speaker recognition. Among various methods proposed in the literature, Bayesian Information Criterion (BIC) based method has been widely used. In this paper, we propose to use a different criterion, Minimum Message Length criterion (MML), which is also well known in the statistical community, on speaker change detection problems. MML is an information theoretic criterion that aims to minimize the message length for the description of both model parameters and the data. Previous studies by Oliver etc. in the area other than speech, showed that MML might be a better criterion than BIC on segmentation problems. We extended their work and applied MML criterion to speaker change detection problems. Experiments were carried out on two different types of speech data, and so far, comparable results between BIC and MML have been obtained.

Full Paper

Bibliographic reference.  Liu, Chaojun / Yan, Yonghong (2000): "Speaker change detection using minimum message length criterion", In ICSLP-2000, vol.3, 514-517.