Rename Dockerfile to dockerfile

Dockerfile → dockerfile

@@ -1,18 +1,22 @@
+# Dockerfile
+
 # Use the official Python slim image
 FROM python:3.10-slim-bookworm
 
-# Set environment variables to control caching and Python behavior
+# Set environment variables for Python, caching, and llama-cpp build
 ENV DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
 ENV PYTHONDONTWRITEBYTECODE=1
 ENV HF_HOME=/app/cache
-ENV NUMBA_CACHE_DIR=/app/cache/numba_cache
-ENV CUDA_VISIBLE_DEVICES=""
+# For GPU builds (NVIDIA), use "-DLLAMA_CUBLAS=on". For CPU-only, remove the CMAKE_ARGS line.
+ENV CMAKE_ARGS="-DLLAMA_CUBLAS=on"
+ENV FORCE_CMAKE=1
 
-# Install essential system dependencies
+# Install essential system dependencies, including cmake for llama-cpp
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    git \
     build-essential \
+    cmake \
+    git \
     espeak-ng \
     libsndfile1 \
     ffmpeg \
@@ -22,41 +26,36 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
 # Set the working directory
 WORKDIR /app
 
-# ---- START: Non-Root User Configuration ----
-# Create a new user 'appuser' with UID 1000 and create their home directory
+# Create a non-root user for security
 RUN useradd --create-home --uid 1000 appuser
-
-# Switch to the non-root user
 USER appuser
-
-# ---- Add this line to fix the PATH issue ----
 ENV PATH="/home/appuser/.local/bin:${PATH}"
 
-# Create the cache directory and set ownership to the new user
-# This needs to run as root, so we'll adjust the Dockerfile structure slightly
-USER root
-RUN mkdir -p /app/cache && chown -R appuser:appuser /app /app/cache
-USER appuser
-# ---- END: Non-Root User Configuration ----
+# Install llama-cpp-python first, using the build arguments
+RUN pip install --no-cache-dir llama-cpp-python
 
-# Copy requirements file and install dependencies as the non-root user
+# Copy and install the rest of the Python dependencies
 COPY --chown=appuser:appuser requirements.txt .
-RUN pip install --no-cache-dir \
-    -r requirements.txt \
-    --index-url https://download.pytorch.org/whl/cpu \
-    --extra-index-url https://pypi.org/simple
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Download the GGUF model from Hugging Face Hub into a versioned models directory
+# This makes the final image self-contained and ready to run
+RUN huggingface-cli download neuphonic/neutts-air-gguf \
+    neutts-air.gguf \
+    --local-dir /app/models \
+    --local-dir-use-symlinks False
 
-# Clone the NeuTTS Air repository
+# Clone the neutts-air library code as specified
 RUN git clone https://github.com/neuphonic/neutts-air.git
 
-# Copy the application code
+# Copy your main application file
 COPY --chown=appuser:appuser app.py .
 
 # Expose the application port
 EXPOSE 7860
 
 # Add a health check to monitor the application's status
-HEALTHCHECK --interval=30s --timeout=10s --start-period=5s --retries=3 \
+HEALTHCHECK --interval=30s --timeout=10s --start-period=30s --retries=3 \
     CMD curl -f http://localhost:7860/health || exit 1
 
 # Set the command to run the application