useful_commands.txt

# To build the image, from src directory.
podman kill whisper-server ; podman rm whisper-server ; podman build -t whisper-server-image .

# To run the server in development mode (map the server.py file into the container so flask can watch for changes)
podman run -d \
        --name whisper-server \
        -v $PWD/server.py:/src/dictation/server.py:Z \
        -p 8009:5000 \
        --security-opt seccomp=unconfined \
        --ipc=host \
        --group-add keep-groups \
        --device=/dev/dri \
        --device=/dev/accel/accel0 \
        whisper-server-image

# To run the server in "release" mode (caveat emptor: Flask still claims it's not a production-ready server).
podman run -d \
        --name whisper-server \
        -p 8009:5000 \
        --security-opt seccomp=unconfined \
        --ipc=host \
        --group-add keep-groups \
        --device=/dev/dri \
        --device=/dev/accel/accel0 \
        whisper-server-image

# Transcribe a file.
curl --data-binary @courtroom.wav -X POST 127.0.0.1:8009/transcribe

# Generate and enable a systemd unit for the container above.
podman generate systemd whisper-server > $HOME/.config/systemd/user/whisper-server.service
systemctl --user daemon-reload
systemctl --user enable whisper-server.service