Voice Liveliness Identification Assisted by Noise Categorization

Voice Activity Detection (VAD) is a very important front end processing in all Speech and Audio processing applications. The performance of most if not all speech/audio processing methods is crucially dependent on the performance of Voice Activity Detection. Voice activity detection (VAD), is to detect the presence of speech in an audio signal degraded by noise, is widely applied in numerous modern speech communication systems. Since speech signals are non-stationary and contain many transient components, it is appropriate to use, perceptual wavelet packet transform (PWPT) as a tool for feature extraction especially in noisy environments. Voice activity detection (VAD) is a process, which can detect speech and non-speech segments from a audio signal. This method combines a noise robust speech processing feature extraction process together with SVM models trained in different background noises for speech/non-speech classification. A multiclass SVM is also used to classify background noises in order to select SVM model for VAD.