Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Li, Hui, Wang, Hongyu, Chen, Zhijin, Sun, Bohan, Li, Bo
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2405.15093
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912018566479872
author Li, Hui
Wang, Hongyu
Chen, Zhijin
Sun, Bohan
Li, Bo
author_facet Li, Hui
Wang, Hongyu
Chen, Zhijin
Sun, Bohan
Li, Bo
contents Singing voice conversion is to convert the source singing voice into the target singing voice except for the content. Currently, flow-based models can complete the task of voice conversion, but they struggle to effectively extract latent variables in the more rhythmically rich and emotionally expressive task of singing voice conversion, while also facing issues with low efficiency in speech processing. In this paper, we propose a high-fidelity flow-based model based on multi-decoupling feature constraints called RASVC, which enhances the capture of vocal details by integrating multiple latent attribute encoders. We also use Multi-stream inverse short-time Fourier transform(MS-iSTFT) to enhance the speed of speech processing by skipping some complicated decoder processing steps. We compare the synthesized singing voice with other models from multiple dimensions, and our proposed model is highly consistent with the current state-of-the-art, with the demo which is available at \url{https://lazycat1119.github.io/RASVC-demo/}.
format Preprint
id arxiv_https___arxiv_org_abs_2405_15093
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Real-Time and Accurate: Zero-shot High-Fidelity Singing Voice Conversion with Multi-Condition Flow Synthesis
Li, Hui
Wang, Hongyu
Chen, Zhijin
Sun, Bohan
Li, Bo
Audio and Speech Processing
Singing voice conversion is to convert the source singing voice into the target singing voice except for the content. Currently, flow-based models can complete the task of voice conversion, but they struggle to effectively extract latent variables in the more rhythmically rich and emotionally expressive task of singing voice conversion, while also facing issues with low efficiency in speech processing. In this paper, we propose a high-fidelity flow-based model based on multi-decoupling feature constraints called RASVC, which enhances the capture of vocal details by integrating multiple latent attribute encoders. We also use Multi-stream inverse short-time Fourier transform(MS-iSTFT) to enhance the speed of speech processing by skipping some complicated decoder processing steps. We compare the synthesized singing voice with other models from multiple dimensions, and our proposed model is highly consistent with the current state-of-the-art, with the demo which is available at \url{https://lazycat1119.github.io/RASVC-demo/}.
title Real-Time and Accurate: Zero-shot High-Fidelity Singing Voice Conversion with Multi-Condition Flow Synthesis
topic Audio and Speech Processing
url https://arxiv.org/abs/2405.15093