refactor(docker): optimize Dockerfiles for multi-stage builds and production deployment

- Implement multi-stage builds for main and speech Dockerfiles
- Reduce image size by using slim base images
- Improve dependency installation with frozen lockfile and production flags
- Add resource constraints and healthcheck to speech service Dockerfile
- Enhance build caching and separation of build/runtime dependencies

Dockerfile

@@ -1,14 +1,14 @@
 # Use Bun as the base image
-FROM oven/bun:1.0.25
+FROM oven/bun:1.0.25 as builder
 
 # Set working directory
 WORKDIR /app
 
-# Copy package files
-COPY package.json ./
+# Copy only package files first for better layer caching
+COPY package.json bun.lockb ./
 
-# Install dependencies
-RUN bun install
+# Install dependencies with production flag
+RUN bun install --frozen-lockfile --production
 
 # Copy source code
 COPY . .
@@ -16,6 +16,19 @@ COPY . .
 # Build TypeScript
 RUN bun run build
 
+# Create production image
+FROM oven/bun:1.0.25-slim
+
+WORKDIR /app
+
+# Copy only the necessary files from builder
+COPY --from=builder /app/dist ./dist
+COPY --from=builder /app/node_modules ./node_modules
+COPY package.json ./
+
+# Set production environment
+ENV NODE_ENV=production
+
 # Expose port
 EXPOSE 4000
 

docker/speech/Dockerfile

@@ -1,41 +1,58 @@
-FROM python:3.10-slim
+# Use Python slim image as builder
+FROM python:3.10-slim as builder
 
-# Install system dependencies
+# Install build dependencies
 RUN apt-get update && apt-get install -y \
     git \
     build-essential \
+    portaudio19-dev \
+    && rm -rf /var/lib/apt/lists/*
+
+# Create and activate virtual environment
+RUN python -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Install Python dependencies with specific versions and CPU-only variants
+RUN pip install --no-cache-dir torch==2.1.2 torchaudio==2.1.2 --index-url https://download.pytorch.org/whl/cpu
+RUN pip install --no-cache-dir faster-whisper==0.10.0 openwakeword==0.4.0 pyaudio==0.2.14 sounddevice==0.4.6
+
+# Create final image
+FROM python:3.10-slim
+
+# Copy virtual environment from builder
+COPY --from=builder /opt/venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Install only runtime dependencies
+RUN apt-get update && apt-get install -y \
     portaudio19-dev \
     python3-pyaudio \
     && rm -rf /var/lib/apt/lists/*
 
-# Install fast-whisper and its dependencies
-RUN pip install --no-cache-dir torch torchaudio --index-url https://download.pytorch.org/whl/cpu
-RUN pip install --no-cache-dir faster-whisper
-
-# Install wake word detection
-RUN pip install --no-cache-dir openwakeword pyaudio sounddevice
-
-# Create directories
-RUN mkdir -p /models /audio
-
-# Download the base model by default
-# The model will be downloaded automatically when first used
-ENV ASR_MODEL=base.en
-ENV ASR_MODEL_PATH=/models
-
-# Create wake word model directory
-# Models will be downloaded automatically when first used
-RUN mkdir -p /models/wake_word
+# Create necessary directories
+RUN mkdir -p /models/wake_word /audio
 
+# Set working directory
 WORKDIR /app
 
 # Copy the wake word detection script
 COPY wake_word_detector.py .
 
 # Set environment variables
-ENV WHISPER_MODEL_PATH=/models
-ENV WAKEWORD_MODEL_PATH=/models/wake_word
-ENV PYTHONUNBUFFERED=1
+ENV WHISPER_MODEL_PATH=/models \
+    WAKEWORD_MODEL_PATH=/models/wake_word \
+    PYTHONUNBUFFERED=1 \
+    ASR_MODEL=base.en \
+    ASR_MODEL_PATH=/models
 
-# Run the wake word detection service
-CMD ["python", "wake_word_detector.py"] 
+# Add resource limits to Python
+ENV PYTHONMALLOC=malloc \
+    MALLOC_TRIM_THRESHOLD_=100000 \
+    PYTHONDEVMODE=1
+
+# Add healthcheck
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+    CMD ps aux | grep '[p]ython' || exit 1
+
+# Run the wake word detection service with resource constraints
+CMD ["python", "-X", "faulthandler", "wake_word_detector.py"]