Enhancement of Noisy Speech Signals for Hearing Aids

摘要

it is very difficult to understand speech signals in presence of background noise for the normal listeners and hearing impaired persons. The human speech and hearing organ is inherently sensitive to interfering noise. Interfering noise decreases speech intelligibility and quality which makes speech communication troublesome and causing human mistakes. Speech signals from the uncontrolled environments may contain degradation components along with the required speech components. Speech enhancement algorithms removes or reduces the noise and improve one or more perceptual aspects of noisy speech most notably quality and intelligibility. This paper reports a performance evaluation of Spectral subtraction algorithm and its modified versions for Hearing aids in different environments such as restaurant, train and Car environments. Clean speech signals are corrupted by background noise respectively multi-talker babble noise, train noise, and car engine noise at four different signal-to-noise ratio levels-2dB, 0dB, 5dB, 10dB. Subjective and objective type evaluation of enhanced speech signals were carried out. The evaluation of intelligibility and quality of enhanced speech is reported for hearing Aids. Section I introduction, section II explains basic block diagram of speech enhancement, explains basic Spectral subtraction speech enhancement algorithm in detail and its various modified versions of spectral subtraction algorithm, section III explains performance evaluation of this speech enhancement algorithms in terms of quality and intelligibility for Hearing aid, section-IV conclusion.