Configuration Parsing Warning:In adapter_config.json: "peft.task_type" must be a string

Emo

Emo-style generation. Covers: emo rock track, jangly clean to distorted guitars, emotional male vocals; pop emo, fast drums, energetic guitars, melodic emotional vocals; and post-hardcore emo, screamed vocals to clean choruses, heavy distorted guitars.

Quick start

  • Base model: AceStep v1.5 Turbo (2B)
  • Trigger word: roti-em00 (must appear in the prompt to activate)
  • Recommended strength: 0.5
  • Tested strength range: 0.3 โ€“ 1.0
  • Recommended steps: 8
  • Recommended guidance: 7.0
  • Recommended shift: 3.0
  • Sample rate: 48000 Hz

Example prompts

  • emo rock track, jangly clean to distorted guitars, emotional male vocals (140 BPM, E minor)
  • pop emo, fast drums, energetic guitars, melodic emotional vocals (160 BPM, G minor)
  • post-hardcore emo, screamed vocals to clean choruses, heavy distorted guitars (145 BPM, C minor)

Training

  • Rank 64, alpha 128, dropout 0.1
  • Target modules: q_proj, k_proj, v_proj, o_proj
  • 500 epochs / 7500 steps, batch 1, optimizer adamw_fused, lr 0.0003, warmup 100
  • Dataset: 15 synthetic tracks, 3379s total, instrumental fraction 0.0

Files

  • emo-v1.safetensors โ€” PEFT LoRA weights (88,130,248 bytes, sha256 f948472c8819021b1b1931b5d46bf44c3ec7a32d15dd57380a16cbbe9ddc5caa)
  • emo-v1.metadata.json โ€” machine-readable sidecar (schema v1)
  • adapter_config.json โ€” PEFT adapter config

Full metadata: see emo-v1.metadata.json.

Downloads last month
9
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support