This work extents the Hidden Markov Chain (HMC) model for the unsupervised segmentation of multicomponent images. Although the vectorial extension of the model is almost straightforward, we are faced to the problem of estimating a mixture of non-Gaussian multidimensional densities. In this work, we adopt an Independent Component Analysis (ICA) approach that allows the mutual dependance between ...

We investigate the problem of acoustic modeling in which prior language-specific knowledge and transcribed data are unavailable. We present an unsupervised model that simultaneously segments the speech, discovers a proper set of sub-word units (e.g., phones) and learns a Hidden Markov Model (HMM) for each induced acoustic unit. Our approach is formulated as a Dirichlet process mixture model in ...

This paper investigates different statistical modeling frameworks for articulatory speech data obtained using real-time (RT) magnetic resonance imaging (MRI). To quantitatively capture the spatio-temporal shaping process of the human vocal tract during speech production a multi-dimensional stream of direct image features is extracted automatically from the MRI recordings. The features are close...

Speaker variability is a well-known problem of state-of-theart Automatic Speech Recognition (ASR) systems. In particular, handling children speech is challenging because of substantial differences in pronunciation of the speech units between adult and child speakers. To build accurate ASR systems for all types of speakers Hidden Markov Models with Gaussian Mixture Densities were intensively use...

% I % % I % Comcmrrs~Accvracy Comcs?~Acnuscy Experimental results: The above adaptation approaches were evaluated with both the dialect adaptation and speaker adaptation experiments using the TIMIT corpus. MLhased stochastic matching [2] of model-space transformation was used as the benchmark for comparisons. The experiments were designed such that phonemes defined in the TIMIT database were re...

This paper describes one of the biometric systems text-independent speaker verification. It discusses the different stages of speaker verification in text-independent systems as well mentioning other systems for speaker verification. Each stage has its subparts, so those parts are discussed as well. The methods for the speaker-verification are displayed in the article. Feature extraction from t...

The popularity of mobile devices offers an ideal platform for personalized recognizers. With data collected from the user, the personalized recognizer with better matched acoustic and linguistic characteristics can offer not only better recognition accuracy but also less computational time. In this paper, we propose a scenario that a small data set (500 utterances with annotation) can be collec...

Although there have been some promising results in computer lipreading, there has been a paucity of data on which to train automatic systems. However the recent emergence of the TCDTIMIT corpus, with around 6000 words, 59 speakers and seven hours of recorded audio-visual speech, allows the deployment of more recent techniques in audio-speech such as Deep Neural Networks (DNNs) and sequence disc...