Start here
What TajweedFormer V4 gives an app, and the short way from sign-in to a working build.
TajweedFormer V4 is a model that hears Quran recitation. It runs inside your app, on the learner's device, and returns what it heard: each sound with its timing and confidence, and separate outputs for lengthening, gemination, ghunna and qalqalah.
There are four SDKs: Android, Apple (iOS and macOS), web, and Windows and Linux. They are free, and the source is public on GitHub.
The short way
- Get access. Sign in with your Hugging Face account on sdk.quranlab.ai, accept the licence and say what you are building. It takes about a minute, and most requests are accepted on the spot.
- Copy one prompt. Your Home has a Copy prompt button. Paste the prompt into your coding assistant, inside your project. It installs the SDK, creates and wires your key, and checks the setup. What the prompt does.
- Run the app once. On its first launch the app registers itself and shows up on your Home.
By hand
Pick your platform. Each page has the install line for the newest SDK, where the key goes, what the platform needs, and the link to the quick start in the package's README.
| Platform | Package | Quick start |
|---|---|---|
| Android | Kotlin, Maven Central | Android |
| iOS and macOS | Swift Package Manager | Apple |
| Browsers and Node | npm, WebAssembly | Web |
| Windows and Linux | C, built from source with CMake | Windows and Linux |
What to read next
- Concepts: an app, a key, a model version, the response time, and what a device downloads.
- Keys: what the key does and where to keep it.
- Recording sharing: how learners can give recordings to Quran Lab, and what you never see.
- Privacy: every field the SDK sends, and what to declare in the App Store and Google Play forms.
Older SDK versions
Quran Lab serves the newest SDK only. An app on an older version gets no model until it updates; its key stays the same. Your app's page says when an update is required and gives you a prompt for your coding assistant that does it.