Documentation
A voice for every line.
Loabi TTS turns text into natural, expressive speech — in ދިވެހި and English. Here's how to use it.
What it is
A neural text-to-speech engine. Type or paste any text and generate clear, human-like audio you can play and download. No setup, no accounts — open the studio and start.
How to use it
1. Type or paste text
Works with Dhivehi (Thaana), English, or a mix of both. The editor auto-detects the script and switches direction.
2. Generate
Hit the button — a single natural voice speaks your text. Audio appears in the player below.
3. Play, download, revisit
Play right from the browser, download as WAV, or pick any earlier generation from the Recent list.
ދިވެހި — built in
Type in Thaana and the editor switches to right-to-left automatically, rendered in a clean Dhivehi typeface. Mixed Dhivehi and English in the same line is fine.
Tips for the best results
- · Punctuation shapes delivery — commas and full stops become natural pauses.
- · Every result can be played, downloaded as a WAV, and reopened from Recent.
Browser extension
Read-aloud, everywhere. Select text on any page, right-click → "Read aloud with Loabi" — no tab-switching, no copy-paste. Synthesis runs entirely on-device (WebGPU/WASM), same engine as the studio above. Model files download once on first use and are cached locally after that.
chrome://extensions,
enable Developer mode (top right), click Load unpacked, and select
the unzipped folder.
about:debugging#/runtime/this-firefox,
click Load Temporary Add-on…, and select manifest.json
inside the unzipped folder. Firefox removes temporary add-ons on restart, so you'll
need to reload it each time you restart the browser.