ElevenLabs
ElevenLabs gives you studio-quality voices for narrating Dina recordings. Setup is just an API key — no SDKs, no Cloud projects, no models to download.
At a glance — Create an API key on ElevenLabs with Text to Speech + Voices permissions, paste it into Dina's AI settings. Total time: ~3 minutes.
Why ElevenLabs
ElevenLabs is the highest-quality TTS provider Dina supports. Use it when you want professional, human-sounding narration on tutorials, walkthroughs, or marketing videos.
For a free, fully on-device alternative, Dina also includes Qwen TTS — switch between providers any time in Settings → AI → Text to Speech.
Step 1 — Open the API keys page
Visit the ElevenLabs developer portal and sign in to your account.
This page lets you create and manage API keys for use in apps like Dina.
Step 2 — Create a new API key
- Click Create API key.
- Give it a name —
Dinais a good default. - Enable both Text to Speech and Voices permissions.
- Click Create, then copy the key (starts with
xi_).
Important — Both permissions are required. Without Text to Speech, narration won't generate; without Voices, Dina can't list the available models for you to pick from.
Step 3 — Paste it into Dina
- Open Dina.
- Press ⌘+, to open Settings.
- Choose the AI tab, then switch to the Text to Speech segment.
- Pick ElevenLabs (API) from the provider dropdown.
- Paste the API key into the API Key field and click Save.
Dina stores the key in your Mac's Keychain — it never leaves your machine. After saving, Dina fetches the available models automatically.
Warning — Don't share your API key with others or commit it to git. If it leaks, return to the developer portal and revoke it; you can mint a new one in seconds.
Step 4 — Pick a model
Once the key is saved, Dina shows the Available Models card with everything your account has access to.
- Click any model row to set it as your default.
- The selected model is used for all narration generated in Dina.
- To re-fetch the list later (for example after enabling new models on ElevenLabs), click the refresh icon at the top of the card.
That's it — narration in Dina now uses ElevenLabs.
Common errors
| Error | Fix |
|---|---|
| "Invalid API key" when saving | The key has a typo or trailing whitespace. Re-copy it from the ElevenLabs developer portal |
| Models list stays empty after saving | The Voices permission isn't enabled on your key. Edit the key on ElevenLabs to grant it |
| Generation fails with "permission denied" | The Text to Speech permission isn't enabled. Edit the key to add it |
| Generated audio is silent | Confirm a model is selected — the chosen one shows a checkmark icon |
| "Quota exceeded" mid-render | You've hit your monthly character limit. Check usage at elevenlabs.io/app/usage |
What's next
- Want a free, offline alternative? Dina includes Qwen TTS — fully on-device, works without an internet connection. Pick Qwen TTS (On-Device) in the provider dropdown to download the models.
- ElevenLabs charges per character. Track your usage at elevenlabs.io/app/usage, and consider switching models — the Flash family is much cheaper than Multilingual v2 with comparable quality for many use cases.
- Switching between providers in Settings → AI is instant; each provider keeps its own credentials independently, so you can flip between them without re-pasting keys.