Podcastfy

Podcastfy is an open-source Python package and CLI, an alternative to NotebookLM's podcast feature, that turns websites, PDFs, images, YouTube videos or a topic into multilingual audio conversations. It uses cloud models with your own API keys, or a local LLM and Edge TTS without keys.

• Inputs: websites, PDFs, images, YouTube videos, text or a topic
• Short (2-5 minutes) or longform (30+ minutes) podcasts
• Transcripts from 100+ LLMs, including OpenAI, Anthropic and Google models, or local LLMs
• Voices from OpenAI, Google, ElevenLabs or Microsoft Edge text to speech
• Custom conversation style, language, structure and voices
• Python package, command line and a FastAPI server in Docker

Tasks

Turn text into speech, Summarize documents & videos

People

Developers building podcast-style audio into apps, Content creators turning articles and videos into audio, Educators and researchers making audio versions of their material

• The default setup needs a Gemini and an OpenAI API key
• Google's multispeaker voices are English only and need extra Google Cloud setup
• You set it up yourself in Python 3.11 or later, with ffmpeg

Is Podcastfy free?

Yes. Podcastfy has a free plan: Open source (Apache-2.0 license): free to run; local LLM and Edge TTS need no keys. We last checked github.com's pricing page on 2026-11-12.

What does the Podcastfy free plan include?

Open source (Apache-2.0 license): free to run; local LLM and Edge TTS need no keys. Source: https://github.com/souzatharsis/podcastfy (checked 2026-11-12).

How do I get Podcastfy for free?

Steps: Install with pip install podcastfy; Set up a local LLM and Edge TTS, or add your API keys; Generate a podcast from your URLs in Python or the CLI. Start at https://github.com/souzatharsis/podcastfy (checked 2026-10-24).

Do I need an account to use Podcastfy for free?

Not for the free option: the free route is to self-host Podcastfy (Install with pip install podcastfy). Checked 2026-10-24.

What's the catch with Podcastfy's free plan?

The default models need Gemini and OpenAI API keys. Google's multispeaker voices are English only. Checked 2026-10-24.

Do I need API keys?

It depends on the models. The default uses Google's Gemini for transcripts and OpenAI for voices, which need API keys. A local LLM with Microsoft Edge TTS needs no API keys.Source: github.com/souzatharsis/podcastfy/blob/main/usage/config.md

What can Podcastfy turn into a podcast?

Websites, PDFs, images, YouTube videos, text and user-provided topics, as short podcasts or longform episodes of 30+ minutes.Source: github.com/souzatharsis/podcastfy

Is Podcastfy an alternative to NotebookLM?

Yes. The maker describes it as an open-source API alternative to NotebookLM's podcast feature, focused on programmatic and customizable generation.Source: github.com/souzatharsis/podcastfy

Free offer

Verified 2026-11-12 on github.com/souzatharsis/podcastfy

…software is licensed under Apache 2.0. See instructions if you would like to use podcastfy in your…

Page content

Checked 2026-10-24 against

• github.com/souzatharsis/podcastfy
• podcastfy.ai
• github.com/souzatharsis/podcastfy/blob/main/usage/config.md

How we check free offers