README

Requirements

• UV (can be installed via curl -LsSf https://astral.sh/uv/install.sh | sh)
• Internet access to download any models (including during runtime)

Setup

• Clone the repo
• uv venv --python 3.11 && source .venv/bin/activate
• uv sync

Running the workflow steps

The steps can be run from within the repo root directory, each step requires the output from the previous step.

Some steps have optional parameters, which will use sensible defaults.

Loading and cleaning data

uv run python src/document_to_podcast/load_data.py \
--input_file "$(pwd)/example_data/Mozilla-Trustworthy_AI.md" \
--output_folder "$(pwd)output"

Generating the podcast script

uv run python src/document_to_podcast/generate_script.py \
--input_file "$(pwd)/output/cleaned.txt" \
--output_folder "$(pwd)/output"

Generating the podcast audio

uv run python src/document_to_podcast/generate_audio.py \
--input_file "$(pwd)/output/podcast.txt" \
--output_folder "$(pwd)/output"

Notes

You can also supply config as a path to a file containing JSON, or as a JSON string.

For example, running the load_data script and supplying the --config flag with the following JSON:

{
  "input_file": "/Users/foo/src/document-to-podcast/example_data/Mozilla-Trustworthy_AI.md",
  "output_folder": "/Users/foo/src/document-to-podcast/output"
}

uv run python src/document_to_podcast/load_data.py \
--config '{"input_file": "/Users/foo/src/document-to-podcast/example_data/Mozilla-Trustworthy_AI.md", "output_folder": "/Users/foo/src/document-to-podcast/output"}'

or if the JSON is stored in a file at /Users/foo/src/document-to-podcast/my_config.json:

uv run python src/document_to_podcast/load_data.py \
--config "/Users/foo/src/document-to-podcast/my_config.json"