我也一样，我希望有大佬能优化H3的音质，甚至能够输出ASMR级别的声音！！！

Thanks for sharing this. I'm struggling to actually make this work completely. It does work - it outputs roughly what I asked for, but not exactly. Maybe it's because I'm using two similar voices. The voices get intermingled even after trying a few variations, changing the order, even swapping out voice samples.

I also continually have issues with snippets of dialog right at the start of clips, with video as well.