Youthful Sonson Sound with a good vocal range / Youth Sound Acheng RVC model

Acheng is the latest gentle gentleman-tone model trained by Miaoyin Workshop. His chat dataset lasts about 30 minutes, with the singing part about 15 minutes. His voice is gentle but not lacking masculinity.

A Cheng's voice can be used for reading texts, and using a gentle tone makes it even more perfect. Especially for ancient-style reading or dubbing, it's truly outstanding.

In singing, if you tap a small breathing noise, you can adjust the breath-proof tear slider. After testing one or two songs, there might be a slight breathing noise; just lower it a bit.

Below, I will show a comparison of timbre before and after reasoning, as well as the performance of singing without a cappella or reverb.

Before reasoning

<!--[if lt IE 9]>document.createElement('audio');<![endif]-->

After reasoning

After reasoning

Singing reasoning (no cappella + no reverb).

Parameters

training log