This week, Slima grew a pair of ears — your draft can be listened to.
Slima Audio — chapters as TTS audiobooks
Click the Listen button at the top of any chapter. The whole chapter becomes an audiobook.
Features:
- Multi-language TTS — Chinese, English, Arabic, Spanish, Korean
- Sync-highlighted captions — the line being read lights up
- Playback bar + speed control (0.5x–2x)
- Markers — drop a marker on any line that feels off, jump back to edit
Why audiobook? Long-form writers have two recurring scenarios — reviewing yesterday’s pages on a commute, or re-hearing your own prose when your eyes have gone numb to it. Slima Audio solves both.
Streaming AI responses
AI conversations switched from “wait three seconds for the whole reply to drop” to token-by-token streaming.
| Dimension | Difference |
|---|---|
| Feel | Watching the AI type is much better than staring at a spinner |
| Save tokens | Spot a wrong direction mid-stream → cancel immediately |
Small UX change. Full-stack rewrite (front and back streaming protocols replaced).
Character avatars + color system
Each character can now have:
- An avatar (upload your own or pick from the preset library)
- An accent color
The avatar shows up in:
- The character card
- The relationship-graph node
- Slima Audio’s per-line speaker tag (great for dialogue-heavy chapters)
- The dialogue block side-marker in the editor
The color system flows automatically into relationship edges, Insight charts, and the upcoming Beta Reader rating cards.
Next week
Slima’s biggest signature feature ships — AI Beta Reader Personas.