Improved single channel phase‐aware speech enhancement technique for low signal‐to‐noise ratio signal

Abstract

In the state-of-the-art single channel speech enhancement techniques, the short-time spectral amplitude is modified while the effect of the phase corruption due to the contamination of additive noise is neglected. This study introduces an improved speech enhancement algorithm based on a phase-aware multi-band spectral subtraction technique which estimates the spectral amplitude of the clean speech signal by considering the phase of the speech and noise signal components, and uses the estimated phase of the clean speech signal for signal reconstruction in the time domain. Experimental results show that the proposed algorithm yields better performance in terms of various objective and composite quality measures and other intelligibility assessment metrics while compared with other existing spectral subtraction methods. Using the composite objective measure quality evaluation technique, it is observed that the overall signal quality of the enhanced speech signal is improved on an average by 70% at 0 dB global input signal-to-noise ratio by using the proposed approach.

Two-Stage Temporal Processing for Single-Channel Speech Enhancement

“…Most of the conventional solutions have been proposed in spectral domain [1] [3][4] [5], where the noise statistics is estimated from the STFT spectrum of the noisy speech. The magnitude spectrum of desired clean speech is estimated by multiplying a frequency dependent spectral gain function with the noisy signal spectrum.…”

Block‐online multi‐channel speech enhancement using deep neural network‐supported relative transfer function estimates

“…Single-channel techniques rely only on spectral filtering, whereas multi-channel methods also exploit the spatial information. Single-channel enhancement is constituted by classical techniques such as spectral subtraction [2,3] or minimum mean square estimator [4], as well as modern techniques based either on statistical principles [5,6] or supported by machine learning (e.g. [7] and overview in [8]).…”

MMSE and maximum a posteriori estimators for speech enhancement in additive noise assuming a t ‐location‐scale clean speech prior

“…Hence, due to noisy phase spectrum, which can have a deteriorating effect on perceptual information [14], the performance of the enhancements algorithms can be reduced specially in low SNR case and fastchanging noise signals. Therefore, some recent studies have presented phase aware speech enhancement methods [14,15].…”