Podcast Generate

To generate a podcast script from web search results, input a specific topic and an output directory. The tool searches the web, synthesizes sources, drafts a dual-host or single-host script using an LLM, and outputs a Markdown file plus a WAV audio file.

Yes, you can convert text documents into dual-host podcast audio. By reading file-based text materials, the system drafts a conversational script and uses TTS synthesis to produce a ready-to-publish WAV audio file alongside a Markdown script.

The maximum duration for TTS podcast audio generation is 20 minutes. The system features automatic duration control ranging from 3 to 20 minutes, ensuring the synthesized WAV audio and script match your specified time constraints.

LLM-based podcast generation does not require manual audio editing. It uses the z-ai-web-dev-sdk for LLM script drafting and TTS synthesis to automatically validate duration and output a final, ready-to-publish WAV audio file.

Podcast script generation from documents supports standard text and document formats for reading input materials. It processes these file-based materials to automatically output a Markdown podcast script and a synthesized WAV audio file.

A limitation of using TTS synthesis for automated podcast production is that duration is capped at 3 to 20 minutes. The system performs duration validation to ensure the final WAV audio adheres to these defined constraints before publishing.