Tomoko Matsui

TL;DR: It is proved that this new family of kernels to handle time series, notably speech data, within the framework of kernel methods which includes popular algorithms such as the support vector machine is positive definite under favorable conditions.

TL;DR: The distortion-intersection measure (DIM), which was introduced as a VQ-distortion measure to increase the robustness against utterance variations, is effective and the speaker identification rates using a continuous ergodic HMM are strongly correlated with the total number of mixtures irrespective of the number of states.

TL;DR: Methods that create models to specify both speaker and phonetic information accurately by using only a small amount of training data for each speaker are investigated and supplementing these methods by adding a phoneme-independent speaker model to make up for the lack of speaker information.

TL;DR: A new method of creating speaker-specific phoneme models consisting of tied-mixture HMMs and adapts the feature space of the tied- mixtures to that of the speaker through phoneme-dependent/independent iterative training is proposed.

TL;DR: This paper compares a VQ (vector quantization)-distortion-based speaker recognition method and discrete/continuous ergodic HMM (hidden Markov model)-based ones, especially from the viewpoint of robustness against utterance variations, and shows that a continuous ergodIC HMM is as robust as a V Q-distortion method when enough data is available.