Podcast Generate

To generate a podcast script from text, the Skill converts your input documents or web-search results into a structured dialogue. It outputs a ready-to-publish podcast_script.md file using natural pacing and dual-host or single-host formats.

Yes, you can convert web search results directly into podcast audio. The Skill processes web-sourced information to generate both a structured podcast_script.md and a final podcast.wav audio file using TTS.

The text-to-speech system auto-adjusts the podcast duration from 3 to 20 minutes based on content length. It enforces output structure, speaker alternation, and validation rules during audio generation.

No extensive formatting is required before generating a podcast. You can provide raw documents or topics, and the Skill will automatically structure the content into a dual-host or single-host dialogue with natural pacing.

Dual-host podcast generation structures content as an alternating dialogue between two speakers, while single-host uses a monologue format. The Skill enforces speaker alternation rules for dual-host and outputs a validated script.

When converting long research documents into a podcast, the main limitation is the auto-adjusted duration cap of 20 minutes. Content exceeding this length will be constrained by the validation rules enforced during script generation.