S$^2$Voice: Style-Aware Autoregressive Modeling with Enhanced Conditioning for Singing Style Conversion
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917211330838528 |
|---|---|
| author | Wang, Ziqian Xia, Xianjun Huang, Chuanzeng Xie, Lei |
| author_facet | Wang, Ziqian Xia, Xianjun Huang, Chuanzeng Xie, Lei |
| contents | We present S$^2$Voice, the winning system of the Singing Voice Conversion Challenge (SVCC) 2025 for both the in-domain and zero-shot singing style conversion tracks. Built on the strong two-stage Vevo baseline, S$^2$Voice advances style control and robustness through several contributions. First, we integrate style embeddings into the autoregressive large language model (AR LLM) via a FiLM-style layer-norm conditioning and a style-aware cross-attention for enhanced fine-grained style modeling. Second, we introduce a global speaker embedding into the flow-matching transformer to improve timbre similarity. Third, we curate a large, high-quality singing corpus via an automated pipeline for web harvesting, vocal separation, and transcript refinement. Finally, we employ a multi-stage training strategy combining supervised fine-tuning (SFT) and direct preference optimization (DPO). Subjective listening tests confirm our system's superior performance: leading in style similarity and singer similarity for Task 1, and across naturalness, style similarity, and singer similarity for Task 2. Ablation studies demonstrate the effectiveness of our contributions in enhancing style fidelity, timbre preservation, and generalization. Audio samples are available~\footnote{https://honee-w.github.io/SVC-Challenge-Demo/}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_13629 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | S$^2$Voice: Style-Aware Autoregressive Modeling with Enhanced Conditioning for Singing Style Conversion Wang, Ziqian Xia, Xianjun Huang, Chuanzeng Xie, Lei Audio and Speech Processing Signal Processing We present S$^2$Voice, the winning system of the Singing Voice Conversion Challenge (SVCC) 2025 for both the in-domain and zero-shot singing style conversion tracks. Built on the strong two-stage Vevo baseline, S$^2$Voice advances style control and robustness through several contributions. First, we integrate style embeddings into the autoregressive large language model (AR LLM) via a FiLM-style layer-norm conditioning and a style-aware cross-attention for enhanced fine-grained style modeling. Second, we introduce a global speaker embedding into the flow-matching transformer to improve timbre similarity. Third, we curate a large, high-quality singing corpus via an automated pipeline for web harvesting, vocal separation, and transcript refinement. Finally, we employ a multi-stage training strategy combining supervised fine-tuning (SFT) and direct preference optimization (DPO). Subjective listening tests confirm our system's superior performance: leading in style similarity and singer similarity for Task 1, and across naturalness, style similarity, and singer similarity for Task 2. Ablation studies demonstrate the effectiveness of our contributions in enhancing style fidelity, timbre preservation, and generalization. Audio samples are available~\footnote{https://honee-w.github.io/SVC-Challenge-Demo/}. |
| title | S$^2$Voice: Style-Aware Autoregressive Modeling with Enhanced Conditioning for Singing Style Conversion |
| topic | Audio and Speech Processing Signal Processing |
| url | https://arxiv.org/abs/2601.13629 |