Experiments with Tree-Structured MMI Encoders on the RM Task

This paper describes the tree-structured maximum mutual information (MMI) encoders used in SSrs Phonetic Engine ® to perform large-vocabulary, continuous speech recognition. The MMI encoders are arranged into a two-stage cascade. At each stage, the encoder is trained to maximize the mutual information between a set of phonetic targets and corresponding codes. After each stage, the codes are compressed into segments. This step expands acousticphonetic context and reduces subsequent computation. We evaluated these MMI encoders by comparing them against a standard minimum distortion (MD) vector quantizer (encoder). Both encoders produced code streams, which were used to train speaker-independent discrete hidden Markov models in a simplified version of the Sphinx system [3]. We used data from the DARPA Resource Management (RM) task. The two-stage cascade of MMI encoders significantly outperforms the standard MD encoder in both speed and accuracy.