SpecTokenizer: A Lightweight Streaming Codec in the Compressed Spectrum Domain
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_ | 1866918169424166912 |
|---|---|
| author | Wan, Zixiang Zhang, Guochang He, Yifeng Wei, Jianqiang |
| author_facet | Wan, Zixiang Zhang, Guochang He, Yifeng Wei, Jianqiang |
| contents | Neural Audio Codecs (NACs) have gained growing attention in recent years as technologies for audio compression and audio representation in speech language models. While mainstream NACs typically require G-level computation and M-level parameters, the performance of lightweight and streaming NACs remains underexplored. This paper proposes SpecTokenizer, a lightweight streaming codec that operates in the compressed spectral domain. Composed solely of alternating CNN and RNN layers, SpecTokenizer achieves greater efficiency and better representational capability through multi-scale modeling in the compressed spectrum domain. At 4 kbps, the proposed SpecTokenizer achieves comparable or superior performance compared to the codec with state-of-the-art lightweight architecture while requiring only 20% of the computation and 10% of the parameters. Furthermore, it significantly outperforms the codec when using similar computational and storage resources. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21209 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SpecTokenizer: A Lightweight Streaming Codec in the Compressed Spectrum Domain Wan, Zixiang Zhang, Guochang He, Yifeng Wei, Jianqiang Audio and Speech Processing Sound Neural Audio Codecs (NACs) have gained growing attention in recent years as technologies for audio compression and audio representation in speech language models. While mainstream NACs typically require G-level computation and M-level parameters, the performance of lightweight and streaming NACs remains underexplored. This paper proposes SpecTokenizer, a lightweight streaming codec that operates in the compressed spectral domain. Composed solely of alternating CNN and RNN layers, SpecTokenizer achieves greater efficiency and better representational capability through multi-scale modeling in the compressed spectrum domain. At 4 kbps, the proposed SpecTokenizer achieves comparable or superior performance compared to the codec with state-of-the-art lightweight architecture while requiring only 20% of the computation and 10% of the parameters. Furthermore, it significantly outperforms the codec when using similar computational and storage resources. |
| title | SpecTokenizer: A Lightweight Streaming Codec in the Compressed Spectrum Domain |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2510.21209 |