Musaix Pro 2026

Musaix Pro is a production-focused AI audio analysis dashboard.

The product direction is now:

Next.js dashboard
+ Supabase auth/database/storage
+ Python FastAPI audio-analysis service
+ Librosa feature extraction
+ Gemini/OpenAI interpretation layer
+ Vercel deploy
+ Cloud Run audio worker

Current Product Path

Musaix Canvas v0.1 remains locked to a small production path:

Reference -> Prompt -> Project -> Export

The audio-analysis layer extends that foundation into real track intelligence:

Upload audio -> Extract metrics -> Store JSON -> Generate AI summary -> Export report

No agent swarm yet. No Pinecone, Chroma, Spotify OAuth, Ableton bridge, Google Drive sync, or multi-agent orchestration until the core analysis loop ships.

Stack

• Next.js App Router
• React
• TypeScript
• Tailwind CSS
• Supabase Auth
• Supabase Postgres
• Supabase Storage
• Vercel AI Gateway
• Python FastAPI
• Librosa / NumPy / SciPy

Routes

• / — launch page
• /canvas — Musaix Canvas
• /api/generate-prompt — structured prompt generation route

Services

• services/audio-api — Python FastAPI service for real audio feature extraction

Environment

NEXT_PUBLIC_SUPABASE_URL=https://xexzplnyzblflucfvbwt.supabase.co
NEXT_PUBLIC_SUPABASE_PUBLISHABLE_KEY=
SUPABASE_SERVICE_ROLE_KEY=
AI_GATEWAY_API_KEY=
MUSAIX_MODEL=openai/gpt-5.5
GOOGLE_GENERATIVE_AI_API_KEY=
OPENAI_API_KEY=
MUSAIX_AUDIO_API_URL=http://localhost:8080
NEXT_PUBLIC_APP_URL=http://localhost:3000

Supabase

Apply the base schema first:

supabase/schema.sql

Then apply the audio-analysis extension:

supabase/audio-analysis-schema.sql

Active Supabase project:

xexzplnyzblflucfvbwt

Core tables:

• projects
• references
• prompts
• exports
• tracks
• analysis_jobs
• analysis_results

Local Dev

Install web dependencies:

npm install
npm run dev

Run the audio API:

cd services/audio-api
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8080

Windows PowerShell:

cd services/audio-api
python -m venv .venv
.\.venv\Scripts\Activate.ps1
pip install -r requirements.txt
uvicorn main:app --reload --port 8080

Build Order

1. Keep /canvas stable.
2. Add Supabase Storage audio bucket.
3. Add upload + persistent player.
4. Call services/audio-api /analyze.
5. Store metrics in analysis_results.
6. Render waveform, spectral, chroma, and metric cards.
7. Generate Gemini/OpenAI summary from verified metrics.
8. Export JSON/PDF reports.