I Built a Local AI Music Studio

Description and use

Miso is a local AI music studio that integrates multiple music models and post‑processing tools, running in Docker containers to provide a unified workflow for generation, editing, and remixing.

Problem

Users want a single interface to experiment with local music models and manage post‑generation tasks such as editing, stem separation, voice conversion, and MIDI transcription.

Outcome

Miso enables local, GPU‑accelerated music generation and post‑processing workflows, reducing the need for multiple command‑line tools and allowing users to experiment with different models and editing techniques while preserving project lineage.

Workflow

• 1. Install Miso via Docker Compose. 2. Select a music model (e.g., ACE‑Step 1.5, YuE2, MiniMax Music 3, HeartMuLa, Stable Audio 3).
• 3. Input a prompt or upload a file. 4. Generate a full song or repaint a selected waveform section.
• 5. Use the stem separation tool to split the track into vocals, drums, bass, etc.
• 6. Transcribe a vocal stem to MIDI notes. 7. Convert MIDI to an ABC score and adjust tempo/key. 8. Feed the score to YuE2 to generate a new arrangement.
• 9. Mix, solo, mute, or replace stems using the built‑in mixer. 10. Export the final take as WAV or MP3.

Evidence

Miso is a self‑hosted local music generation and remix studio built on top of audio.cpp.

Source

The whole thing runs as two Docker containers. One runs Miso itself. The other runs audio.cpp and owns the GPU models.

Source

You can separate a song, transcribe the vocal stem, turn those note events into an ABC melody, inspect the detected tempo and key, correct them if necessary, and feed the score into YuE2 with a completely different style.

Source