Features · Models
The Preview, and the models.
Read this page before you pay.
The Preview.
The Preview synthesises real sound in your browser in plain code. A robot voice from formants, music beds from chords, bass, arpeggio and drums, effects shaped by words, and dialogue with each turn placed. It runs in your browser, so a run queued by someone else is made when the app is open.
- Real files, no model
- No key needed
- Labelled Simulated in the composer
The providers
What is ready where.
| Provider | Makes | State |
|---|---|---|
| Preview | Speech, music, effects, dialogue | Ready, simulated, in your browser |
| Gemini speech | Speech | Connected in code; ready where a key is set |
| ElevenLabs voice | Speech | Connected in code for one voice; ready where a key is set |
| OpenAI speech | Speech (optional) | Connected in code; ready where a key is set |
| Lyria 2 | Music, 30 s | Not connected |
| ElevenLabs effects | Sound effects | Not connected |
Be clear about this
What we do not know for you.
- Pre-betaWhether a real model is on in the hosted app
It needs a key on the server, or in your workspace's Geenius OS secrets. The providers list in the app says what is ready where you are.
Questions
About models.
Is the sound made by an AI model?
Not by default today. The Preview that makes it is simulated in your browser: a robot voice, synthesised beds and word-shaped effects. A real voice model is connected in code and works only where its key is set on the server.
Where does my text go?
With the Preview, it is used in your browser. If a real voice model is on, your text goes to that model's provider.