podcast-generator

Generate speech-to-text audio files from documents using the OpenAI API.

Goal

Build a simple application that:

1. File selection — pick input files (PDF, text, etc.) via a file selector
2. Text extraction — process and prepare the file content for text-to-speech
3. Text-to-speech — send extracted text to an API (OpenAI first) and receive audio output
4. Iterative improvement — refine the pipeline over time

Current State

Phase 1 — Project scaffolding & API integration. The foundation is in place:

• OpenAI SDK installed and ready for text-to-speech API calls
• Development tooling configured (ruff, mypy, pytest, poe)
• Basic test infrastructure exists

Setup

Requires Python 3.11+ and uv.

# Install dependencies
uv sync

# Update lockfile after any dependency changes
uv lock

Development

All tasks are managed via Poe the Poet.

Command	Description
uv run poe dev	Run everything — format, test, lint
uv run poe dev-format	Run ruff linter and formatter
uv run poe dev-test	Run pytest
uv run poe dev-lint	Run mypy type checks

Individual tools

# Lint and format
uv run ruff check .
uv run ruff format .

# Type check
uv run mypy .

# Test
uv run pytest

Project Structure

.
├── main.py            # Entry point
├── pyproject.toml     # Project config, deps, tool settings, Poe tasks
├── uv.lock            # Locked dependency graph
└── tests/
    └── test_main.py   # Smoke test

Tech Stack

Category	Tool
Language	Python 3.11+
Package manager	uv
API client	OpenAI SDK
Linter / formatter	ruff
Type checker	mypy
Testing	pytest
Task runner	poethepoet