Improving Test-Time Performance of RVQ-based Neural Codecs
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_ | 1866914052432723968 |
|---|---|
| author | Kim, Hyeongju Lee, Junhyeok Morton, Jacob Lee, Juheon Yang, Jinhyeok |
| author_facet | Kim, Hyeongju Lee, Junhyeok Morton, Jacob Lee, Juheon Yang, Jinhyeok |
| contents | The residual vector quantization (RVQ) technique plays a central role in recent advances in neural audio codecs. These models effectively synthesize high-fidelity audio from a limited number of codes due to the hierarchical structure among quantization levels. In this paper, we propose an encoding algorithm to further enhance the synthesis quality of RVQ-based neural codecs at test-time. Firstly, we point out the suboptimal nature of quantized vectors generated by conventional methods. We demonstrate that quantization error can be mitigated by selecting a different set of codes. Subsequently, we present our encoding algorithm, designed to identify a set of discrete codes that achieve a lower quantization error. We then apply the proposed method to pre-trained models and evaluate its efficacy using diverse metrics. Our experimental findings validate that our method not only reduces quantization errors, but also improves synthesis quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_19186 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Improving Test-Time Performance of RVQ-based Neural Codecs Kim, Hyeongju Lee, Junhyeok Morton, Jacob Lee, Juheon Yang, Jinhyeok Audio and Speech Processing Sound The residual vector quantization (RVQ) technique plays a central role in recent advances in neural audio codecs. These models effectively synthesize high-fidelity audio from a limited number of codes due to the hierarchical structure among quantization levels. In this paper, we propose an encoding algorithm to further enhance the synthesis quality of RVQ-based neural codecs at test-time. Firstly, we point out the suboptimal nature of quantized vectors generated by conventional methods. We demonstrate that quantization error can be mitigated by selecting a different set of codes. Subsequently, we present our encoding algorithm, designed to identify a set of discrete codes that achieve a lower quantization error. We then apply the proposed method to pre-trained models and evaluate its efficacy using diverse metrics. Our experimental findings validate that our method not only reduces quantization errors, but also improves synthesis quality. |
| title | Improving Test-Time Performance of RVQ-based Neural Codecs |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2509.19186 |