- Home
- AI Podcast Generators
- Podcastfy
Podcastfy
What it is
Podcastfy is an open-source Python package that uses generative AI to turn multimodal material into multilingual audio conversations. It accepts websites, PDFs, images, YouTube videos, text, transcript files, and user-provided topics; topic-based generation can use grounded real-time web search. You can customize format, style, voices, language, and structure, and generate short podcasts of 2–5 minutes or longform podcasts of 30 minutes or more. Transcript generation supports more than 100 LLM models, including OpenAI, Anthropic, and Google models. Listed text-to-speech options include OpenAI, Google, ElevenLabs, and Microsoft Edge. Local LLMs are supported for transcript generation, which the project describes as offering increased privacy and control. Audio exports as MP3. The project documents Python package and command-line use, plus a beta FastAPI deployment for URLs. Requirements include Python 3.11 or higher and ffmpeg. API key needs depend on the selected transcript and audio models; a documented local LLM with Edge TTS setup requires no API keys. The project is licensed under Apache 2.0 and names content creators, educators, researchers, and accessibility advocates as example users.
For and against
For
- Accepts websites, PDFs, images, videos, text, and topics.
- Supports customizable formats, voices, languages, and structure.
- Exports MP3 audio.
- A documented local LLM and Edge TTS setup needs no API keys.
Against
- Requires Python 3.11 or higher and ffmpeg.
- API key needs vary by selected models.
- FastAPI deployment is beta.
- Google multispeaker TTS is limited to English and needs extra setup.
The verdict
Podcastfy offers flexible source inputs, episode lengths, and model choices for generating audio conversations. Getting started requires Python and ffmpeg, and model selection can affect API key requirements.
Who it suits
It suits content creators, educators, researchers, and accessibility advocates who want to turn varied source material into audio conversations. It also fits users comfortable with Python package or command-line workflows.
Plans and prices
Podcastfy open-source Python package
No paid plan or usage allowance is stated on the opened maker pages.
Read 8 Oct 2026 · github.com
On the record
- API key requirements
- API keys depend on the selected transcript and audio models; the configuration guide also lists a Local LLM with Edge TTS setup requiring no API keys.github.com · read 7 Oct 2026
- API keys
- The configuration guide says API key requirements depend on the selected transcript and audio models; its default setup uses Gemini and OpenAI keys.github.com · read 8 Oct 2026
- Customization
- Users can customize conversation format, style, voices, transcript, audio, language, and structure.github.com · read 8 Oct 2026
- Generation from topics
- The project says it can generate podcasts from a topic using grounded real-time web search.github.com · read 8 Oct 2026
- Inputs
- It accepts websites, PDFs, images, YouTube videos, text, and user-provided topics.github.com · read 7 Oct 2026
- Input sources
- It accepts websites, PDFs, images, YouTube videos, text, and user-provided topics as input.github.com · read 8 Oct 2026
- Interfaces
- The project provides Python package and CLI usage, and documents a FastAPI deployment as beta for URLs.github.com · read 7 Oct 2026
- Language models
- The project says transcript generation supports more than 100 models, including OpenAI, Anthropic, and Google models.github.com · read 8 Oct 2026
- License
- The project is licensed under Apache 2.0.github.com · read 7 Oct 2026
- License and audience
- The software is licensed under Apache 2.0, and the README describes use cases for content creators, educators, researchers, and accessibility advocates.github.com · read 8 Oct 2026
- Local models
- Podcastfy supports local LLMs for transcript generation and describes them as providing increased privacy and control.github.com · read 8 Oct 2026
- No-key option
- The configuration guide gives an example combining a local LLM with Edge text to speech that requires no API keys.github.com · read 8 Oct 2026
- Podcast length
- It can generate short podcasts of 2–5 minutes or longform podcasts of 30 minutes or more.github.com · read 8 Oct 2026
- Prerequisites
- The quickstart requires Python 3.11 or higher and ffmpeg for audio processing.github.com · read 8 Oct 2026
- Requirements
- The quickstart lists Python 3.11 or higher and ffmpeg for audio processing as prerequisites.github.com · read 7 Oct 2026
- Text to speech
- Supported text-to-speech options listed include OpenAI, Google, ElevenLabs, and Microsoft Edge.github.com · read 8 Oct 2026
- TTS limitation
- The configuration guide says Google’s multispeaker TTS model is limited to English and needs extra setup.github.com · read 8 Oct 2026
- Ways to use it
- The README documents installation as a Python package, a command-line interface, and a beta FastAPI service for URLs.github.com · read 8 Oct 2026
