Style consistent model improves singing voice quality and timing
SRF-SVB: Style-Consistent Singing Voice Beautifying via Rectified Flow
Sound
Summary
Many amateur singers want to make their singing sound better by fixing pitch and rhythm while keeping their unique voice style. The authors developed a new method called SRF-SVB that enhances singing with high sound quality and speed. It also keeps the singer's style by focusing on their unique voice and expressive patterns. Tests on songs in English and Chinese show that this method works better than existing ones.
What this means in practice
- •For music app developers: Integrate high-quality pitch and rhythm correction while preserving singer style into consumer singing apps.$Commercial implications: Enables improved singing enhancement features for apps targeting casual and amateur singers, delivering superior audio results efficiently.
- •For karaoke system designers: Enhance karaoke machines to correct amateur singing pitch and rhythm without losing the singer’s unique voice characteristics.
Authors
Wenhui Li, Biao Dong, Liwei Hu, Jiqing Han, Yongjun He
Abstract
Singing voice beautifying (SVB) aims to correct pitch and rhythm of amateur singing while enhancing vocal quality, preserving lyrics and the singer's timbre. Existing methods, however, suffer from limited generation quality and efficiency, and tend to neglect the preservation of the singer's style. We propose SRF-SVB, a style-consistent model for SVB via rectified flow, which achieves high-fidelity and efficient beautification covering pitch and rhythm correction. Furthermore, we design a context-guided masked mel-spectrogram inpainting mechanism that effectively preserves the amateur singer's style, including unique timbre and expressive patterns. Experiments on both English and Chinese test sets show that SRF-SVB outperforms baseline models in most objective and subjective metrics.