Ultra-Low-Bitrate Speech Coding with Pretrained Transformers

TL;DR

This paper introduces a neural speech codec using pretrained Transformers combined with a convolutional encoder, achieving high-quality speech synthesis at ultra-low bitrates of 600 bps, outperforming traditional codecs.

Contribution

It presents a novel neural speech codec architecture that leverages pretrained Transformers for improved long-range dependency modeling at very low bitrates.

Findings

Achieves 600 bps speech coding with better quality than original neural codecs.

Subjective evaluations show comparable or superior quality to conventional codecs at higher bitrates.

Transformer-enhanced codec outperforms traditional methods in low-bitrate speech synthesis.

Abstract

Speech coding facilitates the transmission of speech over low-bandwidth networks with minimal distortion. Neural-network based speech codecs have recently demonstrated significant improvements in quality over traditional approaches. While this new generation of codecs is capable of synthesizing high-fidelity speech, their use of recurrent or convolutional layers often restricts their effective receptive fields, which prevents them from compressing speech efficiently. We propose to further reduce the bitrate of neural speech codecs through the use of pretrained Transformers, capable of exploiting long-range dependencies in the input signal due to their inductive bias. As such, we use a pretrained Transformer in tandem with a convolutional encoder, which is trained end-to-end with a quantizer and a generative adversarial net decoder. Our numerical experiments show that supplementing the convolutional encoder of a neural speech codec with Transformer speech embeddings yields a speech codec with a bitrate of $600\,\mathrm{bps}$ that outperforms the original neural speech codec in synthesized speech quality when trained at the same bitrate. Subjective human evaluations suggest that the quality of the resulting codec is comparable or better than that of conventional codecs operating at three to four times the rate.