hugpadhugpad

ACE-Step / acestep-captioner

Audio11B88076Updated Feb 3, 2026
Hugging Face

Downloads

880

Likes

76

Last updated

Feb 3, 2026

License

mit

ACE-Step Captioner is the annotation model used by ACE-Step v1.5 for training data labeling. It is a professional-grade music captioning model that generates detailed, structured descriptions of audio content.

Tasks & library

  • text to audio
  • transformers

Tags

  • transformers
  • qwen2_5_omni
  • text-to-audio
  • music
  • audio

Technical specifications

Parameters11B
Librarytransformers
LanguagesNot available
Licensemit