VoiceStudio Help

1. Open the app and choose the Text-to-Speech tab.
2. Type or paste the text you want spoken.
3. Pick an engine and a voice from the dropdown. If you're not sure, leave the default selected — it works well for most cases and runs fastest on this server.
4. Click Generate and wait a few moments. Longer text takes longer to produce.
5. Play the result right in the browser, or download it as an audio file using the download button next to the player.

This teaches the app to speak in a specific person's voice, using a short sample recording as a reference.

1. Go to the Voice Clone tab.
2. Upload a clean audio sample of the voice — 10–30 seconds of clear speech with no background noise or music works best.
3. Give the voice a name so you can find it again later.
4. Type the new sentence you want that voice to say.
5. Click Generate. The cloned voice will be saved for reuse next time.

This server uses a single access key to protect administrative functions. Treat it like a password: don't share it, paste it into chat messages, or post it anywhere public. If you're just asked to try the site casually, use the share link/PIN you were given instead of the main key.

This site is currently reachable from the public internet. That's a deliberate, temporary setup while we're getting it running — it will later be restricted to trusted devices only.

Generation is slow.

This server processes audio using its CPU rather than a graphics card, so generation — especially longer clips or voice cloning — can take noticeably longer than you may have seen elsewhere. This is expected.

I got logged out / it's asking for a key again.

Browser sessions expire after a period of inactivity for security. Just re-enter the access key you were given.

My download didn't start.

Try right-clicking the play button and choosing "Save audio as...", or reload the page and generate again.

Something looks broken.

Reload the page first. If it still doesn't work, note what you were doing and let the site owner know.