Paste text or choose a .txt / .pdf file. PDFs are read on your machine.
.txt
.pdf
Pick a chip (deep-dive, brief, critique, debate) or type a custom briefing. The writer turns the source into a script.
The writer is the LLM (Gemini, any OpenAI-compatible endpoint, or Ollama). The voice speaks each line (edge is free; OpenAI-compatible uses /v1/audio/speech; Gemini is native). Each speaker can use a different voice.
/v1/audio/speech
Write the script, edit any line, then “Speak this script”. The episode joins the in-memory gallery — Save writes the MP3 + a JSON sidecar to your podcasts folder.
No account, no telemetry. The only network traffic is to the writer and voice providers you configure. The gallery lives in memory and disappears when the server stops.
Close