PhoenixCodec: Taming Neural Speech Coding for Extreme Low-Resource Scenarios
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911462149062656 |
|---|---|
| author | Wan, Zixiang Zhao, Haoran Zhang, Guochang Han, Runqiang Wei, Jianqiang Zou, Yuexian |
| author_facet | Wan, Zixiang Zhao, Haoran Zhang, Guochang Han, Runqiang Wei, Jianqiang Zou, Yuexian |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21196 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PhoenixCodec: Taming Neural Speech Coding for Extreme Low-Resource Scenarios Wan, Zixiang Zhao, Haoran Zhang, Guochang Han, Runqiang Wei, Jianqiang Zou, Yuexian Audio and Speech Processing Sound 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. |
| title | PhoenixCodec: Taming Neural Speech Coding for Extreme Low-Resource Scenarios |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2510.21196 |