🧠 Neuroacoustic MVP — Technical Scope

Goal: Build a system that generates personalised, neuroscience-informed sound experiences (Focus / Calm / Sleep / High) using AI-generated music and entrainment DSP, personalised via user inputs or wearables.

Duration: 4–6 weeks

Stack: Node.js (API) · React (frontend) · Python (DSP) · ElevenLabs Music API · Docker · AWS/Supabase

1️⃣ MVP OBJECTIVES

🎵 AI-generated base music for each desired state using ElevenLabs Music API.

🧬 Neuroscience entrainment layer (binaural or isochronic beats) blended over music.

⚙️ Real-time adjustment of entrainment parameters (frequency, intensity).

📊 User session tracking (selected state, time, completion).

💡 Simple adaptive loop using HRV or self-feedback.

💻 Modern web interface (React) to select a state and play generated sound.

2️⃣ USER FLOW (MVP version)

User opens web app → selects mental state (Focus / Calm / Sleep / High).

Frontend calls Node.js API → requests base music generation from ElevenLabs.

Node receives music → sends to Python DSP microservice to add neuro layer.

Node returns final track URL (S3/CDN or data URI) → React plays audio.

User gives optional feedback (e.g. “felt calmer”, “too intense”) → stored to DB.

Optimiser adjusts parameters for next session.

3️⃣ CORE FEATURES (Phase 1 MVP)

Feature	Description	Owner
State Selector UI	Simple React interface (4 buttons: Focus, Calm, Sleep, High)	Frontend
AI Music Generation	Generate music from ElevenLabs Music API via Node	Backend
DSP Processor (Entrainment)	Python microservice adds binaural or isochronic beats based on preset frequency	DSP
Adaptive Parameters	Backend logic adjusts diffHz & mix level based on state or user feedback	Node
Session Storage	Save user ID, state, timestamp, settings, feedback	Backend (Supabase)
Web Player	HTML5/React audio player with progress and loop	Frontend
Admin Dashboard (optional)	Simple session list (user ID, state, feedback)	Backend

4️⃣ TECH STACK

🟢 Frontend (React)

React + Vite + Tailwind

Web Audio API for volume control and optional local modulation

Socket.io-client for live updates (future)

Deployed on Vercel / Netlify

🟣 Backend (Node.js)

Express or Fastify REST API

Routes:

POST /session → generate and return track URL
POST /feedback → log user feedback

Integrations:

ElevenLabs Music API
DSP microservice (HTTP)
Supabase (DB + auth)

Deployed on AWS ECS or Railway

🔵 DSP Microservice (Python)

FastAPI + Pydub/Numpy

Endpoint /render

Input: base music URL, diffHz, mixDb, mode
Output: mixed track URL or data URI

Default modes:

Focus: 10 Hz (alpha)
Calm: 7 Hz (theta)
Sleep: 4 Hz (delta)
High: 16 Hz (beta)

🟠 Database

Supabase (PostgreSQL + API)

Tables:

users (id, name, preferences)
sessions (id, user_id, state, params, result_url, timestamp)
feedback (session_id, rating, notes)

⚫ Optional Analytics

Simple metrics: total sessions, average duration, state popularity

Logged via Supabase or Google Analytics

5️⃣ API CONTRACTS (simplified)

🎵 `POST /session`

Request:

json
{
  "userId": "123",
  "state": "focus",
  "durationSec": 120
}

Response:

json
{
  "sessionId": "abc123",
  "trackUrl": "https://cdn.anomate.ai/tracks/focus_abc123.wav",
  "parameters": { "mode": "binaural", "diffHz": 10, "mixDb": -18 }
}

💬 `POST /feedback`

Request:

json
{
  "sessionId": "abc123",
  "rating": 4,
  "comment": "Felt very focused after 3 min"
}

Response:

json
{ "ok": true }

6️⃣ TIMELINE (4–6 weeks)

Week	Deliverables	Owner
1	Setup repo, Docker, ElevenLabs API, baseline React UI	Dev Lead
2	Implement /session route + DSP service (static presets)	Backend + DSP
3	Integrate audio playback + basic feedback form	Frontend
4	Deploy on staging (AWS/Vercel) + DB connection	DevOps
5	Add adaptive parameter tuning (simple rule-based)	Backend
6	QA, user testing, polish UX, deploy MVP	All

7️⃣ MVP OUTPUT EXAMPLE

User selects “Focus”

→ System generates music with 10 Hz binaural beat at -18 dB mix

→ Plays in browser

→ After session, user reports “good focus”

→ Optimiser slightly increases duration next time

8️⃣ NEXT PHASE (POST-MVP)

Once MVP is validated:

Add WHOOP / Muse / Apple Health integrations for HRV, EEG.

Personalise entrainment dynamically (Bayesian optimiser).

Add user profile AI that remembers what worked best.

Mobile app wrapper (React Native).

Paid subscriptions and session library.

⚠️ 9️⃣ Notes / Constraints

ElevenLabs currently outputs full tracks; ensure API plan supports desired call rate.

Keep entrainment under -18 dB to avoid auditory fatigue.

Include disclaimer: “For relaxation & focus purposes only, not medical treatment.”

Implement volume guard and session timeout (max 30 min).

✅ 10️⃣ Deliverable Summary

Deliverable	Description
🎛️ Web app (React)	Select state, play session, give feedback
🔗 Node.js API	Handle sessions, call ElevenLabs, integrate DSP
🧠 Python DSP	Add binaural/isochronic modulation
💾 Database	Store sessions & feedback
🌐 Hosted demo	Accessible MVP URL
📘 Documentation	Setup + API spec + state parameter table

MVP Pitch (1)System Architecture Overview (1)Fine Tune Audience by Feedback (1)

Neuroacoustic MVP — Technical Scope