CQNV: A combination of coarsely quantized bitstream and neural vocoder for low rate speech coding

TL;DR

This paper introduces CQNV, a low-bitrate speech coding framework combining coarsely quantized traditional parameters with a neural vocoder, achieving higher quality speech reconstruction at 1.1 kbps compared to existing methods.

Contribution

The paper presents a novel CQNV framework that integrates coarsely quantized parameters with a neural vocoder and adds a parameters processing module to enhance speech quality at low bitrates.

Findings

Achieves higher speech quality at 1.1 kbps than Lyra and Encodec at 3 kbps.

Demonstrates effectiveness through subjective and objective evaluations.

Reduces bitrate while maintaining or improving speech quality.

Abstract

Recently, speech codecs based on neural networks have proven to perform better than traditional methods. However, redundancy in traditional parameter quantization is visible within the codec architecture of combining the traditional codec with the neural vocoder. In this paper, we propose a novel framework named CQNV, which combines the coarsely quantized parameters of a traditional parametric codec to reduce the bitrate with a neural vocoder to improve the quality of the decoded speech. Furthermore, we introduce a parameters processing module into the neural vocoder to enhance the application of the bitstream of traditional speech coding parameters to the neural vocoder, further improving the reconstructed speech's quality. In the experiments, both subjective and objective evaluations demonstrate the effectiveness of the proposed CQNV framework. Specifically, our proposed method can achieve higher quality reconstructed speech at 1.1 kbps than Lyra and Encodec at 3 kbps.