We present in this paper a comparative study between several spectral subtraction methods for speech enhancement, with the aim of improving the performance of LPC-based low bit rate speech coders in noisy environments. We also describe an improved voice-activity detector which is able to better distinguish between unvoiced speech and noise at low SNRs. Incorporating the studied spectral subtraction methods in ACELP4800 and LPC10-E resulted in significantly improved performances. The studied algorithms are compared in terms of the LPC spectral distance between the uncorrupted and enhanced speech signals.
Cite as: Floc'h, A.L., Salami, R., Mouy, B., Adoul, J.-P. (1992) Evaluation of linear and non-linear spectral subtraction methods for enhancing noisy speech. Proc. ETRW on Speech Processing in Adverse Conditions, 131-134
@inproceedings{floch92_spac,
author={A. Le Floc'h and R. Salami and B. Mouy and J-P. Adoul},
title={{Evaluation of linear and non-linear spectral subtraction methods for enhancing noisy speech}},
year=1992,
booktitle={Proc. ETRW on Speech Processing in Adverse Conditions},
pages={131--134}
}