Phone deactivation pruning in large vocabulary continuous speech recognition

Abstract

Introduces a new pruning strategy for large vocabulary continuous speech recognition based on direct estimates of local posterior phone probabilities. This approach is well suited to hybrid connectionist/hidden Markov model systems. Experiments on the Wall Street Journal task using a 20000 word vocabulary and a trigram language model have demonstrated that phone deactivation pruning can increase the speed of recognition-time search by up to a factor of 10, with a relative increase in error rate of less than 2%.

Related Papers

• → A dynamic language model for speech recognition(1991)126 cited
• → Class phrase models for language modeling(2002)45 cited
• → Speech recognition using a stochastic language model integrating local and global constraints(1994)9 cited
• → Decoupling Word-Pair Distance and Co-occurrence Information for Effective Long History Context Language Modeling(2015)2 cited
• → Improving language models by using distant information(2007)3 cited