Save a sound or voice clip as a MiniMax H3 RefMod and reuse it in your AI videos. Audio references, no training needed.
MiniMax H3Audio → Reference packAdvanced
A RefMod replaces LoRA training for MiniMax H3. Instead of training a LoRA, you save a reusable reference of a character, a video or a sound in a few minutes, then load it whenever you generate.
The RefMod pack has four workflows: three that create RefMods (from images, video or audio) and the main MiniMax H3 RefMod workflow that uses them. This workflow turns an audio clip, like a voice, music or ambience, into an audio RefMod.
Great for:
Giving an AI influencer a recurring voice or sound
Consistent music or ambience across videos
Sound-driven character clips
How it works
Upload your audio (the first 30 seconds are used).
Hugging Face