2 commits

Author	SHA1	Message	Date
John Lightner	45f6863131	feat(01-02): add generation tuning params, seed control, multi-take, and run logging ... Tuning session revealed strength=0.3 as optimal default for cover mode. Added --caption, --noise-strength, --guidance, --steps, --shift, --sampler, --vel-clamp, --vel-ema, --seed, --takes flags. Each output now has a JSON sidecar logging all parameters for reproducibility.	2026-04-11 03:24:27 -05:00
John Lightner	5a233898c8	feat(01-01): create hum2inst.py CLI pipeline script ... - Single-file CLI wrapping ACE-Step XL-SFT cover mode - argparse with --instrument, --output, --strength, --duration flags - Auto-detect input WAV duration via torchaudio - Caption templates for piano, guitar, saxophone, violin, flute - CUDA GPU check with clear error message - Silence detection on output audio - User-friendly output naming with instrument + timestamp	2026-04-11 02:11:53 -05:00