Dockerfile

# ----------- START Dockerfile -----------
# Use an official Python runtime as a parent image
FROM python:3.10

# Set the working directory in the container
WORKDIR /app

# Set environment variables
# - PYTHONUNBUFFERED: Ensures Python output is sent straight to terminal without buffering
# - HF_HOME: Specifies where Hugging Face models/datasets should be cached INSIDE the container
ENV PYTHONUNBUFFERED=1 \
    HF_HOME=/app/hf_cache

# Install system dependencies
# - ffmpeg: Required by pydub
# - build-essential: May be needed for compiling some Python dependencies
# - git: Sometimes needed if pip installs from git repositories
RUN apt-get update && apt-get install -y --no-install-recommends \
    ffmpeg \
    build-essential \
    git \
    && apt-get clean \
    && rm -rf /var/lib/apt/lists/*

# Copy the requirements file into the container
COPY requirements.txt .

# Install Python dependencies
# Upgrade pip first, use --no-cache-dir to reduce image size
RUN pip install --no-cache-dir --upgrade pip && \
    pip install --no-cache-dir -r requirements.txt

# Copy the model download script BEFORE the main app code
COPY download_models.py .

# Run the model download script during the build process
# This downloads models into the HF_HOME directory defined above
# NOTE: This step can take a LONG time and requires network access during build
RUN python download_models.py

# Copy the rest of the application code into the container
COPY app.py .

EXPOSE 7860
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

# ----------- END Dockerfile -----------