
Podcast Generation
Turns written text into a two-host podcast MP3 with a transcript
- Repository
- bytedance/deer-flow
- Stars
- 84,876
- Licence
- MIT
- Files
- 3 files
What it does
Once installed, the agent can take an article, report or document and write it as a casual conversation between a male and a female host, then run a bundled Python script that calls a text-to-speech service to produce an MP3 podcast and a markdown transcript. It is built around a 'Hello Deer' greeting and a target of roughly 10 minutes of dialogue, and supports English and Chinese. Note that the script depends on external speech services (the code references MiniMax and Volcengine) selected through environment variables.
Great for
- Creators who want a quick audio version of a written article or report
- Students who prefer listening to a summary of study material
- Teams turning internal documentation into an audio briefing
Skip it if
- Content that must be exact or legally verified, since the agent rewrites material into dialogue and may simplify or add errors
- Confidential documents, because the text is sent to third-party speech APIs over the network
- Users without API keys or accounts for the chosen TTS provider, since the script cannot produce audio without them
Try asking
- “Generate a podcast about the history of artificial intelligence from this article”
- “Turn this technical documentation into a 10-minute two-host episode in Chinese”
- “Make an audio explainer of our quarterly report with a transcript I can read”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 3 original files, the MIT licence and an attribution note.
podcast-generation.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip podcast-generation.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip podcast-generation.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Provide clean source text; the agent turns it into dialogue, so check the transcript for accuracy before sharing.
- Specify the language (English or Chinese) explicitly if the content is mixed.
- Confirm the command succeeded before sharing the MP3, since the skill itself warns that existing files do not prove success.
What's inside
- scripts/generate.py
- SKILL.md
- templates/tech-explainer.md
Redistributed unmodified from bytedance/deer-flow under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.