ISCA Archive Eurospeech 1997
ISCA Archive Eurospeech 1997

Wideband coding of speech using neural network gain adaptation

Cheung-Fat Chan, Man-Tak Chu

In this paper, a high-quality wideband speech coder is proposed. The coding structure resembles a LD-CELP coder, however, several novel improvements are made. The gain adapter for the stochastic codebook is driven by a neural network and it updates the excitation gain in a sample-by-sample fashion. The purpose of incorporating a neural network is to exploit both the intra- and inter-frame correlation of speech signal in a non-linear manner. A psychoacoustic model instead of a simple perceptual weighting filter is used to shape the quantization noise. Simulation result shows that the proposed coder can achieve transparent coding of wideband speech at 16 kbps.


doi: 10.21437/Eurospeech.1997-435

Cite as: Chan, C.-F., Chu, M.-T. (1997) Wideband coding of speech using neural network gain adaptation. Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997), 1507-1510, doi: 10.21437/Eurospeech.1997-435

@inproceedings{chan97_eurospeech,
  author={Cheung-Fat Chan and Man-Tak Chu},
  title={{Wideband coding of speech using neural network gain adaptation}},
  year=1997,
  booktitle={Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997)},
  pages={1507--1510},
  doi={10.21437/Eurospeech.1997-435},
  issn={1018-4074}
}