Single-Channel Speech Enhancement Based on Subband Spectral Entropy

Wei Y, Zeng Y, Li C. Single-Channel Speech Enhancement Based on Subband Spectral Entropy. J Audio Eng Soc. 2018;66(3):100-113. doi:10.17743/jaes.2018.0003

@article{Wei2018_19379,
  author = {Wei, Yi and Zeng, Yumin and Li, Chen},
  title = {{Single-Channel Speech Enhancement Based on Subband Spectral Entropy}},
  journal = {Journal of the Audio Engineering Society},
  volume = {66},
  number = {3},
  pages = {100--113},
  year = {2018},
  month = mar,
  publisher = {Audio Engineering Society},
  doi = {10.17743/jaes.2018.0003},
  url = {https://doi.org/10.17743/jaes.2018.0003}
}

TY  - JOUR
TI  - Single-Channel Speech Enhancement Based on Subband Spectral Entropy
AU  - Wei, Yi
AU  - Zeng, Yumin
AU  - Li, Chen
T2  - Journal of the Audio Engineering Society
J2  - J. Audio Eng. Soc.
VL  - 66
IS  - 3
SP  - 100
EP  - 113
PY  - 2018
DA  - 2018/03/06
DO  - 10.17743/jaes.2018.0003
UR  - https://doi.org/10.17743/jaes.2018.0003
PB  - Audio Engineering Society
LA  - en
AB  - The goal of speech enhancement is to make speech more pleasant and understandable, improving one or more perceptual aspects of speech, such as quality or intelligibility. This paper addresses single-channel speech enhancement. The authors explore improved multiband spectral subtraction based on the equivalent rectangular bandwidth (ERB) scale. In the proposed algorithm, the full speech spectrum is divided into different nonuniform frequency bands, and spectral subtraction is performed separately in each band. Moreover, subband spectral entropy is used directly to do the noise estimation rather than using speech endpoint detection. The ERB scale is adopted in the subband spectral entropy instead of the traditional linear scale or the Bark scale. The subband spectral entropy based on ERB scale can obtain a more accurate noise estimation, which can achieve better single-channel speech enhancement. The speech spectrograms, objective measures, and informal subjective listening tests show that the remnant noise is suppressed more by the proposed algorithm than by the Upadhyay’s algorithm.
ER  -