A Noise Robust Speech Activity Detection Algorithm for Voice Activated Hands-Free

H. Bagur V (India)


Speech Detection, Speech Recognition.


This paper proposes an efficient and robust speech starting and end point detection method, which improves the performance of the speech recognition engines in a noisy environment. This proposed method designs a series of speech/non-speech classifiers for speech activity detection. The proposed method uses multiple features of speech for robust speech detection under noisy conditions, especially in a cellular phone or in an automotive environment. The speech/non-speech classification is done using a novel and a robust algorithm called “Adaptive-Thresholding” algorithm. The key advantages of this method are its simple implementation and its low computational complexity. The proposed algorithm is used for the isolated word recognition in a discrete speech recognition system. The performance of the proposed algorithm is measured in a simulated noisy environment with speech wave files recorded under noisy conditions.

Important Links:

Go Back