Efficient speech recognition using subvector quantization and discrete-mixture HMMS

This paper introduces a new form of observation distributions for hidden Markov models (HMMs), combining subvector quantization and mixtures of discrete distrib utions. Despite what is generally believed, we show that discrete-distrib ution HMMs can outperform continuous-density HMMs at significantly faster decoding speeds. Performance of the discrete HMMs is impro ved by using product-code vector quantization (VQ) and mixtures of discrete distributions. The decoding speed of the discrete HMMs is also improved by quantizing subvectors of coefficients, since this reduces the number of table lookups needed to compute the output probabilities. We present efficient training and decoding algorithms for the discrete-mixture HMMs (DMHMMs). Our e xperimental results in the air-travel information domain show that the high level of recognition accuracy of continuous-mixture-density HMMs (CDHMMs) can be maintained at significantly faster decoding speeds. Moreover, we show that when the sam...

• Type: Journal
• Year: 2000
• Where: CSL
• Authors: Vassilios Digalakis, S. Tsakalidis, Costas Harizakis, Leonardo Neumeyer
• Keywords: Automated Reasoning, Continuous-density Hmms, CSL 2000, Discrete Hmms, Ution Hmms