# Any tips how to make Minimax H3 voice consistent? — retained record

> Retained reddit record from reddit:StableDiffusion.

Canonical: https://highsignal.app/data/records/d5648842070865bd

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:StableDiffusion)  published 2026-09-23 04:49 ingested 2026-09-24 02:39

[original source ↗](https://www.reddit.com/r/StableDiffusion/comments/1wnwd43/any_tips_how_to_make_minimax_h3_voice_consistent/)

## retained text



```
So I’m making a 50-second clip using separate 10-second segments, but the problem is that the voice isn’t consistent.

Are there any tips to make the voice more consistent, or would it be better to use QwenTTS instead?
```



document fetched 2026-09-24 00:35 · canonical https://reddit.com/r/StableDiffusion/comments/1wnwd43/any_tips_how_to_make_minimax_h3_voice_consistent
