First episode
Assumes you followed Getting started and finished account init.
1. Confirm AI services
Open Settings → AI services (or check OPENAI_* in .env):
- API key / base URL work
- Chat / ASR / TTS model IDs match your provider
- Image model optional (covers)
Or call MCP get_system_health. Partition map: Settings.
2. (Optional) Persona & voice
| Setting | Role |
|---|---|
| Persona | Who hosts, who they talk to, show vibe |
| Voice | Default TTS; presets or Voice Design |
| Prompts | Cover / spoken / rewrite / flashcard templates |
Defaults are fine for a first run.
3. Drop content in
Home / create supports:
- URL — public articles, direct media links
- Local files — video / audio / text
- Plain text — notes, meeting minutes
Optional: Source plugin, album, per-episode persona / voice / locale.
4. Wait for the pipeline
text
Ingest → extract audio → ASR → spoken script
→ cover / flashcards → TTS → library- Progress on the home job list
- Retry or re-run from a step (skip completed)
- Text inputs skip ASR
Details: Pipeline.
5. Listen & organize
- Job detail: transcript, script, notes, flashcards, cover, audio
- Library: immersive player (progress, speed, sleep timer)
- Albums / star map: themes
6. Level up
Common snags
| Symptom | Check |
|---|---|
| Stuck on ASR | ASR model / key; decodable media |
| Thin script | Source too short; persona / prompt |
| No audio | TTS model & voice name; job error |
| URL fetch fails | Login wall / non-direct → Source plugin or local upload |
More: FAQ.