Codec-independent sound activity detection based on the entropy with adaptive noise update

The existing voice activity detectors (VAD) always depend on specific audio codecs and give the degraded performance in the existence of music signals. This paper presents a sound activity detection method independent of audio codecs. An entropy feature set with adaptive noise estimation update is proposed to improve the performance of the entropy in detecting both speech and music. Afterwards, a codec-independent sound activity detection system is constructed through integrating SNR-based features with the proposed entropy. The experiment shows that the system gives a promising detection accuracy especially for music-only or interlaced music/speech signals compared with AMR-WB VAD performance. The complexity of the system is less than 3 wmops, which is applicable to various kinds of audio codecs. In addition, the system can be easily adjusted to various transmission environments by operation point setting.