PhoenixCodec: Taming Neural Speech Coding for Extreme Low-Resource Scenarios

TL;DR

PhoenixCodec is a neural speech coding framework optimized for extremely low-resource scenarios, achieving high efficiency and robustness with innovative training and architecture strategies.

Contribution

It introduces an integrated low-resource neural speech coding system with CCR training, asymmetric architecture, and noise-invariant fine-tuning, outperforming existing methods.

Findings

Ranked third in LRAC 2025 Challenge Track 1

Achieved best performance at 1 kbps in noisy and reverberant conditions

Demonstrated high intelligibility in clean tests

Abstract

This paper presents PhoenixCodec, a comprehensive neural speech coding and decoding framework designed for extremely low-resource conditions. The proposed system integrates an optimized asymmetric frequency-time architecture, a Cyclical Calibration and Refinement (CCR) training strategy, and a noise-invariant fine-tuning procedure. Under stringent constraints - computation below 700 MFLOPs, latency less than 30 ms, and dual-rate support at 1 kbps and 6 kbps - existing methods face a trade-off between efficiency and quality. PhoenixCodec addresses these challenges by alleviating the resource scattering of conventional decoders, employing CCR to enhance optimization stability, and enhancing robustness through noisy-sample fine-tuning. In the LRAC 2025 Challenge Track 1, the proposed system ranked third overall and demonstrated the best performance at 1 kbps in both real-world noise and reverberation and intelligibility in clean tests, confirming its effectiveness.