Segregation of stop consonants from acoustic interference

Speech segregation from acoustic interference is a very challenging task. Previous systems have dealt with voiced speech with success, but they cannot handle unvoiced speech. We study the segregation of stop consonants, which contain significant unvoiced signals. We propose a novel method that employs onset as a major cue to segregate stop consonants. Our system first detects stops through onset detection and Bayesian classification of acoustic-phonetic features, and then performs grouping based on onset coincidence. The system has been tested and performs well on utterances mixed with various types of interference.