added docker

.dockerignore

@@ -0,0 +1,36 @@
+# Version control
+.git
+.gitignore
+
+# Desktop/native (not needed for Docker)
+tauri/
+apps/
+landing/
+
+# Python artifacts
+backend/venv/
+backend/__pycache__/
+backend/**/__pycache__/
+backend/*.pyc
+backend/*.spec
+backend/data/
+
+# Node artifacts
+node_modules/
+app/node_modules/
+web/node_modules/
+web/dist/
+
+# Build artifacts
+dist/
+*.egg-info/
+
+# IDE/OS
+.vscode/
+.idea/
+.DS_Store
+
+# Docs
+docs/
+*.md
+!backend/requirements*.txt

CHANGELOG.md

@@ -62,6 +62,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   - Self-documenting help system with `make help`
   - Colored output for better readability
   - Supports parallel development server execution
+- **Docker Support** - Dockerfile and docker-compose.yml for containerized deployment
+  - Three compose configurations: full stack, backend-only, and web-only
+  - NVIDIA GPU support with optional CUDA acceleration
+  - Persistent volumes for data and HuggingFace model cache
+  - Health check endpoint integration
 
 ### Changed
 - **README** - Added Makefile reference and updated Quick Start with Makefile-based setup instructions alongside manual setup

Dockerfile

@@ -0,0 +1,31 @@
+FROM pytorch/pytorch:2.6.0-cuda12.6-cudnn9-runtime
+
+# System dependencies:
+#   git        - required for pip install from GitHub (qwen-tts)
+#   libsndfile1 - required by soundfile (audio I/O)
+#   ffmpeg     - required by librosa (audio processing)
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    git \
+    libsndfile1 \
+    ffmpeg \
+    && rm -rf /var/lib/apt/lists/*
+
+WORKDIR /app
+
+ENV PYTHONUNBUFFERED=1
+ENV HF_HOME=/root/.cache/huggingface
+
+# Install Python dependencies
+COPY backend/requirements.txt backend/requirements.txt
+RUN pip install --no-cache-dir -r backend/requirements.txt \
+    && pip install --no-cache-dir git+https://github.com/QwenLM/Qwen3-TTS.git
+
+# Copy backend source
+COPY backend/ backend/
+
+# Data volume mount point
+RUN mkdir -p /app/data
+
+EXPOSE 8000
+
+ENTRYPOINT ["python", "-m", "backend.main", "--host", "0.0.0.0", "--port", "8000", "--data-dir", "/app/data"]

Dockerfile.web

@@ -0,0 +1,34 @@
+# Stage 1: Build the web frontend
+FROM oven/bun:1 AS build
+
+WORKDIR /app
+
+# Copy workspace root config
+COPY package.json bun.lock ./
+
+# Copy the workspaces needed for the build
+COPY app/ app/
+COPY web/ web/
+
+# Create stub package.json for workspaces we don't need but bun requires
+RUN mkdir -p tauri && echo '{"name":"@voicebox/tauri","version":"0.0.0","private":true}' > tauri/package.json
+RUN mkdir -p landing && echo '{"name":"@voicebox/landing","version":"0.0.0","private":true}' > landing/package.json
+
+# Install dependencies
+RUN bun install
+
+# Build the web frontend (skip tsc type checking — Vite transpiles independently)
+RUN cd web && bunx vite build
+
+# Stage 2: Serve with Nginx
+FROM nginx:stable-alpine
+
+# Copy built assets
+COPY --from=build /app/web/dist /usr/share/nginx/html
+
+# Copy Nginx config and entrypoint
+COPY docker/nginx.conf /etc/nginx/conf.d/default.conf
+COPY docker/web-entrypoint.sh /docker-entrypoint.d/40-backend-url.sh
+RUN chmod +x /docker-entrypoint.d/40-backend-url.sh
+
+EXPOSE 3000

README.md

@@ -89,6 +89,79 @@ Voicebox is available now for macOS and Windows.
 
 ---
 
+## Docker
+
+Run Voicebox in Docker — no Python, Bun, or system dependencies required.
+
+### Quick Start
+
+```bash
+# Clone and start both backend + web UI
+git clone https://github.com/jamiepine/voicebox.git
+cd voicebox
+docker compose up -d
+```
+
+- **API**: http://localhost:8000 (Swagger docs at http://localhost:8000/docs)
+- **Web UI**: http://localhost:3000
+
+Verify the backend is running:
+
+```bash
+curl http://localhost:8000/health
+```
+
+### Compose Configurations
+
+| Configuration | Command | Description |
+|---------------|---------|-------------|
+| **Full stack** | `docker compose up -d` | Backend + Web UI |
+| **Backend only** | `docker compose -f docker-compose.backend.yml up -d` | API server only |
+| **Web only** | `BACKEND_URL=http://my-server:8000 docker compose -f docker-compose.web.yml up -d` | Frontend pointing to existing backend |
+
+### GPU Support
+
+For NVIDIA GPU acceleration, install the [NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html), then uncomment the `deploy` section in `docker-compose.yml`:
+
+```yaml
+deploy:
+  resources:
+    reservations:
+      devices:
+        - driver: nvidia
+          count: all
+          capabilities: [gpu]
+```
+
+The backend auto-detects GPU availability at runtime — it works on CPU without any changes.
+
+### Data Persistence
+
+| Volume | Purpose |
+|--------|---------|
+| `voicebox-data` | Database, voice profiles, and generated audio |
+| `huggingface-cache` | Downloaded ML models (Qwen3-TTS, Whisper) |
+
+Data persists across `docker compose down` / `docker compose up` cycles.
+
+### Connecting Clients
+
+**Desktop app** — Use Remote Server mode and point to `http://<host>:8000`.
+
+**API** — The full REST API is available at port 8000:
+
+```bash
+# List voice profiles
+curl http://localhost:8000/profiles
+
+# Generate speech
+curl -X POST http://localhost:8000/generate \
+  -H "Content-Type: application/json" \
+  -d '{"text": "Hello from Docker", "profile_id": "your-profile-id", "language": "en"}'
+```
+
+---
+
 ## Features
 
 ### Voice Cloning with Qwen3-TTS

docker-compose.backend.yml

@@ -0,0 +1,27 @@
+services:
+  backend:
+    build: .
+    ports:
+      - "8000:8000"
+    volumes:
+      - voicebox-data:/app/data
+      - huggingface-cache:/root/.cache/huggingface
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    # Uncomment the following to enable NVIDIA GPU support:
+    # deploy:
+    #   resources:
+    #     reservations:
+    #       devices:
+    #         - driver: nvidia
+    #           count: all
+    #           capabilities: [gpu]
+
+volumes:
+  voicebox-data:
+  huggingface-cache:

docker-compose.web.yml

@@ -0,0 +1,10 @@
+services:
+  web:
+    build:
+      context: .
+      dockerfile: Dockerfile.web
+    ports:
+      - "3000:3000"
+    environment:
+      - BACKEND_URL=${BACKEND_URL:-http://localhost:8000}
+    restart: unless-stopped

docker-compose.yml

@@ -0,0 +1,40 @@
+services:
+  backend:
+    build: .
+    ports:
+      - "8000:8000"
+    volumes:
+      - voicebox-data:/app/data
+      - huggingface-cache:/root/.cache/huggingface
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    # Uncomment the following to enable NVIDIA GPU support:
+    # deploy:
+    #   resources:
+    #     reservations:
+    #       devices:
+    #         - driver: nvidia
+    #           count: all
+    #           capabilities: [gpu]
+
+  web:
+    build:
+      context: .
+      dockerfile: Dockerfile.web
+    ports:
+      - "3000:3000"
+    environment:
+      - BACKEND_URL=http://localhost:8000
+    depends_on:
+      backend:
+        condition: service_healthy
+    restart: unless-stopped
+
+volumes:
+  voicebox-data:
+  huggingface-cache:

docker/nginx.conf

@@ -0,0 +1,14 @@
+server {
+    listen 3000;
+    root /usr/share/nginx/html;
+    index index.html;
+
+    location /assets/ {
+        expires 1y;
+        add_header Cache-Control "public, immutable";
+    }
+
+    location / {
+        try_files $uri $uri/ /index.html;
+    }
+}

docker/web-entrypoint.sh

@@ -0,0 +1,6 @@
+#!/bin/sh
+# Replace the hardcoded dev backend URL with the configured BACKEND_URL
+# in all built JS assets at container startup.
+BACKEND_URL="${BACKEND_URL:-http://localhost:8000}"
+find /usr/share/nginx/html/assets -name '*.js' \
+  -exec sed -i "s|http://localhost:17493|${BACKEND_URL}|g" {} +