SPEECH ENHANCEMENT USING HARMONIC EMPHASIS AND ADAPTIVE COMB FILTERING
The enhancement of single-channel speech degraded by additive noise has been extensively studied in the past and remains a challenging problem because only the noisy speech is available. Techniques have been proposed in the literature to exploit the harmonic structure of voiced speech for enhancing speech quality. In this design, a method that enhances the harmonics of voiced speech without ascribing to any underlying speech models is proposed. The harmonic speech structure obtained through short-time Fourier analysis is enhanced by applying a combination of time and frequency domain-based criteria, which are applicable for white as well as for colored additive noise conditions.
Our method addresses the voiced harmonics specifically, instead of using general vector subspaces theory or typically formulated using Karhunen–Loeve transforms (KLT) solutions. In contrast to many state-of-the-art approaches, the proposed algorithm allows an admissible level of residual noise in the enhanced speech. Since in many real-world applications, complete removal of the degrading noise is neither feasible nor desirable, retaining a low-level background noise yields better perceptual quality. The proposed method improves speech quality by suppressing the noise in the frequency domain with the use of a spectral weighting function. Two design parameters are introduced into the proposed suppression gain, namely the frequency-dependent noise-flooring parameter (FDNFP) and the gain factor. The FDNFP shapes the residual noise in the frequency domain such that the harmonic structure of clean speech is preserved. To further enhance the harmonics of voiced speech, adaptive comb filtering is performed using the gain factor by picking the harmonic peaks from the noisy speech spectrum. Therefore, the proposed algorithm extracts and enhances the harmonics by operating in both the time and frequency domains. The performance of the enhancement method was evaluated by the modified bark spectral distance (MBSD), ITU-Perceptual Evaluation of Speech Quality (PESQ) scores.
Reference Paper: Speech Enhancement Using Harmonic Emphasis and Adaptive Comb Filtering
Author’s Name: Wen Jin, Xin Liu, Michael S. Scordilis, and Lu Han
Source: IEEE
Year: 2010
Request source code for academic purpose, fill REQUEST FORM or contact +91 7904568456 by WhatsApp or info@verilogcourseteam.com, fee applicable.
SIMULATION VIDEO DEMO
