Dockerfile

# UFPAlign Brazilian Portuguese Speech Alignment Container
FROM kaldiasr/kaldi:latest

LABEL maintainer="UFPAlign Hugging Face Space"
LABEL description="UFPAlign - Brazilian Portuguese Forced Phonetic Alignment Tool"
LABEL version="1.0"

# Set environment variables
ENV UFPALIGN_DIR=/opt/UFPAlign
ENV KALDI_ROOT=/opt/kaldi
ENV LC_ALL=pt_BR.UTF-8
ENV LANG=pt_BR.UTF-8
ENV PYTHONPATH=/opt/UFPAlign:$PYTHONPATH

# Update system and install dependencies
RUN apt-get update && \
    apt-get install -y --no-install-recommends \
    sudo \
    curl \
    wget \
    default-jdk \
    locales \
    python3-pip \
    python3-dev \
    python3-setuptools \
    build-essential \
    sox \
    ffmpeg \
    git \
    unzip && \
    # Configure locale for Portuguese (Brazil)
    sed -i '/pt_BR.UTF-8/s/^# //g' /etc/locale.gen && \
    locale-gen && \
    # Upgrade pip
    python3 -m pip install --upgrade pip && \
    # Cleanup
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

# Create UFPAlign directory
RUN mkdir -p ${UFPALIGN_DIR}

# Clone UFPAlign repository
RUN cd /opt && \
    git clone https://github.com/falabrasil/ufpalign.git UFPAlign && \
    cd UFPAlign && \
    # Make the shell script executable
    chmod +x ufpalign.sh

# Install Python dependencies for UFPAlign and FastAPI
RUN pip install --no-cache-dir \
    fastapi \
    uvicorn \
    python-multipart \
    pydantic \
    textgrid \
    pandas \
    numpy \
    scikit-learn \
    scipy \
    matplotlib

# Install additional UFPAlign Python requirements if they exist
RUN cd ${UFPALIGN_DIR} && \
    if [ -f requirements.txt ]; then pip install -r requirements.txt; fi

# Copy application files
COPY app.py /app/app.py

# Create necessary directories
RUN mkdir -p /app/uploads /app/output /app/logs

# Set working directory
WORKDIR /app

# Make sure UFPAlign models and dictionaries are available
RUN cd ${UFPALIGN_DIR} && \
    # Check if demo files exist and are accessible
    ls -la demo/ || echo "Demo directory not found" && \
    # Ensure scripts are executable
    find . -name "*.sh" -exec chmod +x {} \;

# Expose port
EXPOSE 7860

# Health check
HEALTHCHECK --interval=30s --timeout=30s --start-period=5s --retries=3 \
    CMD curl -f http://localhost:7860/health || exit 1

# Start the application
CMD ["python3", "app.py"]