An Improved Speech Endpoint Detection Based on Spectral Subtraction and Adaptive Sub-band Spectral Entropy

摘要

Endpoint detection in strong noise environment plays an important role in speech recognition. This paper presents an improved method of endpoint detection based on the product of spectral subtraction and adaptive sub-band spectral entropy. Firstly, the additive noises are removed by spectral subtraction. Then, background noise estimated value is updated timely. Finally, improved adaptive sub-band spectral entropy is used to detect the endpoints for the enhanced speech. Experimental results show that the method has higher accuracy than traditional methods. Furthermore, for low signal-to-noise ratio, the proposed one has better robustness for different types of noise.