readbeyond/aeneas with Debian Trixie - Currently Python 3.13

Dockerfile

# Use Debian Trixie - Currently Python 3.13
FROM debian:trixie-slim

ENV DEBIAN_FRONTEND=noninteractive

# 1. Fix Shell Encoding Warning
# This fixes the [WARN] you saw in diagnostics
ENV PYTHONIOENCODING=UTF-8
ENV LANG=C.UTF-8
ENV LC_ALL=C.UTF-8

# 2. Install System Dependencies
# python3-dev is REQUIRED for headers
# libespeak-ng-dev is REQUIRED for linking
RUN apt-get update && apt-get install -y \
    python3 \
    python3-venv \
    python3-pip \
    python3-dev \
    build-essential \
    git \
    ffmpeg \
    espeak-ng \
    libespeak-ng-dev \
    festival \
    festival-dev \
    && rm -rf /var/lib/apt/lists/*

# 3. Setup Virtual Environment
ENV VIRTUAL_ENV=/opt/venv
RUN python3 -m venv $VIRTUAL_ENV
ENV PATH="$VIRTUAL_ENV/bin:$PATH"

# 4. Upgrade build tools
RUN pip install --no-cache-dir --upgrade pip setuptools wheel packaging numpy

WORKDIR /tmp/aeneas_build

# 5. Clone the Python 3.12+ compatible fork
RUN git clone -b py312_support https://github.com/avinashvarna/aeneas.git .

# 6. FIX: Symlink Headers for Espeak
RUN ln -s /usr/include/espeak-ng /usr/include/espeak

# 7. FIX: Patch setup.py for Linker (espeak -> espeak-ng)
# This was the magic fix for "ld: cannot find -lespeak"
RUN sed -i 's/"espeak"/"espeak-ng"/g' setup.py

# 8. Install Aeneas
# (We don't force Festival here because CEW/Espeak is working and sufficient)
RUN pip install .

# 9. Cleanup
WORKDIR /app
RUN rm -rf /tmp/aeneas_build

# 10. Verify
RUN python -m aeneas.diagnostics

CMD ["/bin/bash"]

wiki

docker build -t aeneas-final .

docker run --rm \
    -e PYTHONIOENCODING=UTF-8 \
    -v $(pwd):/data \
    aeneas-final \
    python -m aeneas.tools.execute_task \
    /data/audio.wav \
    /data/text.txt \
    "task_language=eng|is_text_type=plain|os_task_file_format=vtt" \
    /data/alignment.vtt

or tweaked:

docker run --rm \
    -e PYTHONIOENCODING=UTF-8 \
    -v $(pwd):/data \
    aeneas-final \
    python -m aeneas.tools.execute_task \
    /data/audio.wav \
    /data/text.txt \
    "task_language=eng|is_text_type=plain|os_task_file_format=vtt|os_task_head_tail_format=hidden|task_adjust_boundary_algorithm=percent" \
    /data/alignment.vtt