Voice and Noise Detection with AdaBoost

Speech recognition is one of our most effective communication tools when it comes to a hands-free (human-machine) interface. Most current speech recognition systems are capable of achieving good performance in clean acoustic environments. However, these systems require the user to turn the microphone on/off to capture voices only. Also, in hands-free environments, degradation in speech recognition performance increases significantly because the speech signal may be corrupted by a wide variety of sources, including background noise and reverberation. Sudden and short-period noises also affect the performance of a speech recognition system. Figure 1 shows a speech wave overlapped by a sudden noise (a telephone call). To recognize the speech data correctly, noise reduction or model adaptation to the sudden noise is required. However, it is difficult to remove such noises because we do not know where the noise overlapped and what the noise was. Many studies have been conducted on non-stationary noise reduction in a single channel (A. Betkowska, et al., 2006), (V. Barreaud, et al., 2003), (M. Fujimoto & S. Nakamura, 2005). But it is difficult for these methods to track sudden noises.