Studio

Studio is where you write scripts and turn them into finished takes.

1. Write

Type or paste into the large editor. Your draft is saved automatically, so it's still there next time you open VoxStudio.

The footer shows the word count and an estimate of how long it will run, which adjusts to your chosen pace.

Punctuation shapes delivery: commas add a breath, an ellipsis… adds a pause, and an exclamation mark lifts the energy!

2. Pick a voice

In the right-hand panel:

1. Choose an engine: System, Kokoro or Chatterbox. Only installed engines appear.
2. Pick a voice from the list. Voices are grouped by language, with yours first. Your voices (cloned) are at the top when Chatterbox is selected.
3. With many voices, use the search box to filter by name or language.

The orb at the top of the panel shows the current voice, and dances while a take plays.

3. Shape the delivery

• Pace: 0.6× to 1.6×.
• Emotion: from Calm to Intense. Available with Chatterbox; greyed out for engines that don't support it.

4. Generate

Click Generate or press ⌘↩.

• Quick engines (System, Kokoro) with short text play almost immediately.
• Chatterbox and long scripts render in the background: the status shows Rendering 40%, and Generate becomes Cancel. Progress also appears in Activity.

5. Work with takes

Each take appears as a card under the editor:

• ▶ / ❚❚ plays and pauses. Click anywhere on the waveform to jump to that point.
• ☆ stars a take you like. Hover to reveal it.
• ⤓ Save as… saves the take as a WAV file wherever you choose.

Tips

• Chatterbox's first take after launch is slow while the model loads (about 45 s on an M1 Pro). After that it's a few seconds per sentence.
• Change where Chatterbox runs in Settings → Performance → Compute device.