Downloads · 30 days
0
mozi1924/mozi-v2
mozi-v2 is a machine learning model from mozi1924. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as gpl-3.0.
Downloads · 30 days
0
Access
Public
Updated Jan 5, 2026
Repo size
368 MB
Likes
0
Public
Click a slice to open those files.
.index311 MB · 84%
From the Hugging Face model README
Mozi V2 是作者 (Mozi) 在 15岁 时训练的 RVC 声音模型。 这是一个物理层面上已经绝版的声音。由于作者已经经历了变声期(生理结构改变),目前的本人(以及后续的 Mozi-3/4 模型)已无法复刻此版本特有的少年音色和高频爆发力。
Mozi V2 is an RVC voice model trained by the author (Mozi) at the age of 15. This is a biologically discontinued voice. Since the author has gone through voice change, this specific juvenile timbre and high-frequency explosive power cannot be reproduced by the current author (nor the later Mozi-3/4 models).
这个模型以其在翻唱 The Marías 的《Hush》中的表现而存盘(Archive)。
"Hush, don't speak..." ➡️ (Screams in horror) ➡️ "...when you spit your venom." —— The legendary "Hush" cover workflow.
此仓库包含以下文件,请下载后放入 RVC 的 weights 和 logs 目录使用:
mozi_v2.pth: 主模型权重文件。added_*.index: 特征检索文件(推荐使用,以还原当年的独特听感)。本模型基于 RVC (Retrieval-based Voice Conversion) 架构。
rmvpe 以获得最佳稳定性,或者 harvest 以获得更浓厚的色彩。0.5 - 0.7。如果是为了复刻尖叫效果,可以适当调高。本模型的原始训练数据已开源,遵循 GPL-3.0 协议: 👉 HuggingFace: mozi1924/mozi-v2
如果你想研究如何炼出这种音色,或者想尝试用更先进的算法(如 RVC v3, Sovits 等)重新训练,欢迎下载数据集。
本项目遵循 GNU General Public License v3.0 (GPL-3.0)。
Time flies, but the voice remains. Enjoy the scream.