podcast-generator

Script generation from research ingests articles or URLs to create a podcast script featuring host/guest dynamics and a strong narrative arc, which is then rendered into multi-voice audio using ElevenLabs text-to-speech.

Yes, you can generate a solo explainer podcast script instead of a multi-voice dialogue by selecting your preferred podcast format, which also supports conversational duo, interview, debate, or narrative structures tailored to your content type.

Yes, you need an ElevenLabs API key to produce text-to-speech podcast audio, as the Skill relies on the ElevenLabs dependency to convert scripted dialogue lines into multi-voice audio before applying loudness normalization via ffmpeg-normalize.

The best way to normalize loudness for text-to-speech podcast episodes is using the integrated ffmpeg-normalize dependency, which processes the rendered ElevenLabs audio chunks to ensure the final output is a listenable, loudness-normalized audio file.

For converting technical papers into audio explainers, conversational duo or interview formats work best, transforming dense research into 10–20 minute episodes with conversational explanations, hooks, and clear takeaways to improve comprehension.