feat: add Dockerfile with CPU-only torch, .dockerignore, and requirements-cpu.txt

- Multi-stage Dockerfile with CPU-only PyTorch
- .dockerignore to exclude build artifacts
- requirements-cpu.txt for CPU-only dependencies
- Health check endpoint for container monitoring

.dockerignore

@@ -0,0 +1,17 @@
+__pycache__/
+*.pyc
+*.pyo
+.env
+.venv/
+venv/
+*.egg-info/
+dist/
+build/
+*.egg
+response_*.wav
+*.log
+.git/
+.gitignore
+.dockerignore
+Dockerfile
+requirements-cpu.txt

Dockerfile

@@ -1,19 +1,13 @@
-ARG BASE_IMAGE=nvidia/cuda:12.2.2-runtime-ubuntu22.04
+FROM python:3.11-slim
 
-FROM ${BASE_IMAGE}
-
-# Set environment variables
 ENV PYTHONUNBUFFERED=1 \
     PYTHONDONTWRITEBYTECODE=1 \
     DEBIAN_FRONTEND=noninteractive \
-    MODEL_CACHE_DIR=/models
+    MODEL_CACHE_DIR=/models \
+    TORCH_CUDA_ARCH_LIST=""
 
-# Install Python and system dependencies
+# Install system dependencies
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    python3.10 \
-    python3.10-venv \
-    python3-pip \
-    build-essential \
     git \
     curl \
     wget \
@@ -22,11 +16,15 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
     ffmpeg \
     && rm -rf /var/lib/apt/lists/*
 
+# Install torch CPU-only first (before other deps that might pull CUDA)
+RUN pip install --no-cache-dir torch torchaudio --index-url https://download.pytorch.org/whl/cpu
+
 # Set up Python virtual environment
 WORKDIR /app
+
+# Copy requirements and install Python deps
 COPY requirements.txt .
-RUN pip install --no-cache-dir --upgrade pip && \
+RUN pip install --no-cache-dir -r requirements.txt
-    pip install --no-cache-dir -r requirements.txt
 
 # Copy application code
 COPY . .
@@ -36,11 +34,14 @@ RUN mkdir -p ${MODEL_CACHE_DIR}
 
 # Set environment variables for models
 ENV TRANSFORMERS_CACHE=${MODEL_CACHE_DIR} \
-    HF_HOME=${MODEL_CACHE_DIR} \
+    HF_HOME=${MODEL_CACHE_DIR}
-    PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True
 
 # Expose WebSocket port
 EXPOSE 8000
 
+# Health check
+HEALTHCHECK --interval=30s --timeout=10s --start-period=60s --retries=3 \
+    CMD curl -f http://localhost:8000/docs || exit 1
+
 # Run the application
 CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

requirements-cpu.txt

@@ -0,0 +1,26 @@
+# WebSocket server
+fastapi
+uvicorn[standard]
+websockets
+webrtcvad
+
+# Speech-to-Text
+faster-whisper
+soundfile
+
+# LLM (CPU-only)
+transformers
+accelerate
+bitsandbytes
+
+# TTS
+torchaudio
+
+# Audio processing
+numpy
+scipy
+
+# Utilities
+python-dotenv
+pydantic
+pydantic-settings