Evaluation of a Multi-Resolution Dyadic Wavelet Transform Method for usable Speech Detection
This addresses the issue of co-channel speech interference in speech communication and speaker identification systems, but it appears incremental as it builds on existing wavelet transform methods.
The paper tackled the problem of detecting usable speech segments in co-channel speech for speaker identification, achieving 95.76% correct detection with 29.65% false alarms on the TIMIT database.
Many applications of speech communication and speaker identification suffer from the problem of co-channel speech. This paper deals with a multi-resolution dyadic wavelet transform method for usable segments of co-channel speech detection that could be processed by a speaker identification system. Evaluation of this method is performed on TIMIT database referring to the Target to Interferer Ratio measure. Co-channel speech is constructed by mixing all possible gender speakers. Results do not show much difference for different mixtures. For the overall mixtures 95.76% of usable speech is correctly detected with false alarms of 29.65%.