podcast-generation

To convert text into a two-host podcast, this Skill generates a structured dialogue script in JSON format and synthesizes it via text-to-speech, producing a finished MP3 file. It transforms articles or reports into engaging audio with clearly defined lines for each host.

You can generate podcast audio from both English and Chinese text content. The Skill supports multilingual audiences by processing articles or reports in either language to produce the final MP3 and optional transcript.

To synthesize text-to-speech into a podcast MP3, you need a structured script JSON, a valid SKILL.md frontmatter, authentication for the TTS service, and the requests dependency installed to run the generator script.

To start generating a podcast from a script JSON, run the podcast generator with the --script-file and --output-file arguments in a single command. You can also include --transcript-file to output an optional readable transcript.

The best way to repurpose blog posts and meeting notes into audio is using a text-to-speech podcast generator that converts written content into a two-host dialogue script and mixes it into a ready-to-publish MP3 file at scale.

Yes, the podcast audio generation process outputs an optional readable transcript alongside the final MP3 file. You can enable this by specifying the --transcript-file argument when running the generator script.