Noise robust voice detector for speaker recognition

The effect of additive noise in a speaker recognition system is well known to be a crucial problem in real life applications. In a speaker recognition system, if the test utterance is corrupted by any type of noise, the performance of the system notoriously degrades. The use of a noise robust voice detector to determine which audio frame is a voice frame is proposed in this work. The new detector is implemented using an Ada-Boosting algorithm in a voiced-unvoiced sound classifier based on speech waveform only. Results reflect better performance of robust speaker recognition based on selected voice segments, respects to unselected segments, in front to additive white noise.

• Type: Conference
• Year: 2008
• Where: ICPR
• Authors: Gabriel Hernández, José R. Calvo, Rafael Fernández, Ivis Rodés, Rafael Martínez
• Keywords: Computer Vision, ICPR 2008, Noise Robust Voice, Speaker Recognition, Speaker Recognition System