Quran speech recognition · in the browser · no server

A Quran model that listens,
and keeps it on the phone.

Naml follows a recitation word by word. Three small streaming models, about 17 MB of weights each, run in any modern browser on our WebAssembly core. Take the model for free, buy the tracker that runs recite.nabilmh.com, or put the Tarawih on your masjid's screen.

every demo below is the real engine's output, replayed

01 The model · free

Three models, streaming as you recite.

Press play. Every 0.32 s each model reads the newest audio (with 0.16 s of look-ahead) and writes what it heard. The timings are what each step took on one CPU core of our server.

Naml Tajweed
77 tajweed phoneme units: madd 2 / 4 / 6, shadda, ghunna, ikhfa, iqlab
Naml Letters
Arabic letters, 8 layers, the earliest word timings
Naml Letters Plus
letters, 10 layers, fewer slips, a little slower

Cached streaming FastConformer, 8 or 10 blocks, ternary weights (−1 / 0 / +1 per 128-input block, 2 bits each). Base model: NVIDIA NeMo FastConformer, CC BY 4.0.

naml · streaming output

Loading the demo…

Try it live

Run Naml Tajweed on your own microphone, in this tab. It downloads about 23 MB once (weights, front end, the wasm core). Nothing is uploaded.

Open weights

The models

Free
  • All three models, weights and wasm core
  • Unit lists, model card, how to run it
  • CC BY 4.0, credit Fanah Solutions and NVIDIA
On Hugging Face soon
02 The tracker · $500

It knows the word you are on.

The tracker turns the model's stream into a place in the mushaf: the word you are reciting, highlighted on the Madani page, the page turned when you reach the end. Start anywhere; it finds you.

  • Live word highlighting on the 15-line Madani mushaf
  • Finds your place in the whole Quran, then follows
  • Page turning, Drive view, Prayer view (follow the imam)
  • Hifz test: a random ayah, you continue, the app checks
  • English of the ayah (Saheeh International)
recite · mushaf view

Loading the demo…

Source code

Tracker code

$500 one-time
  • The full source of recite.nabilmh.com
  • Naml Tajweed engine, tracker, mushaf, views
  • Use it in your own apps and sites
Try the live app first
03 Tajweed correction · coming soon

Mistakes, word by word, a second after.

Each completed word is force-aligned on Naml Tajweed's frames and compared with the likely mistake (a missing shadda, a clear noon for ikhfa, ط read as ت). Strong findings are errors, weaker ones "uncertain". They land in a Mistake History with your own audio to replay.

Measured at the default setting (llr ≥ 8)
False alarms on correct professional recitation0.16 / min (EveryAyah, 73 min) · 0.20 / min (long QUD recordings, 196 min)
Detected, altered-text testshadda 92% · ikhfa 94% · iqlab 97% · haraka 79–94% · ط as ت 90% · ع as ء 96% · ح as ه 88% · ص as س 60%
Cost~0.3–0.6 ms per word on a server CPU

Altered-text figures are optimistic (the text context helps). Madd length and ghunna duration are not detected yet: they are off, and we do not claim them.

recite · mistake history

Loading the demo…

Coming soon

Tajweed correction

$1,000 one-time, planned
  • Everything in the tracker code
  • The tajweed analyser and Mistake History screen
  • The calibration script, to tune it on your audio

One email when it is ready. Nothing else.

04 Masjid kit · $500

Tarawih, followed on screen.

Put a laptop or a TV by the musalla: the ayah the imam is reciting, the word he is on and its English, page and ayah progress. It listens in Sync mode from Al-Fatiha, as the app's Prayer view does, and moves with him from surah to surah.

  • The ayah, the word and its English, large, for a TV or projector
  • Follows the imam from Al-Fatiha into the night's passage
  • Runs on one laptop, offline: no subscription, no per-minute fee
  • Setup guide: audio from the mixing desk, kiosk mode, the screen
recite · prayer view

Loading the demo…

For masjids

Masjid kit

$500 one-time
  • The Tarawih screen, ready for a big display
  • The app's source, set up for one laptop
  • Step-by-step setup guide; your own recording as a test
Prefer we set it up? Ask us
Ways in

Pick the door that fits.

01

Open model

Free
  • Naml Tajweed, Letters, Letters Plus
  • Weights, wasm core, unit lists
  • CC BY 4.0
Hugging Face soon
02

Tracker code

$500 one-time
  • Source of recite.nabilmh.com
  • Highlighting, place finding, page turns
  • Drive, Prayer and Hifz test views
03

Masjid kit

$500 one-time
  • Tarawih on the masjid's screen
  • Ayah, word and English, live
  • Offline, one laptop, setup guide
04

Built for you

Ask
  • We put it into your app or masjid screen
  • Quoted per project
  • Talk to the person who built it
Ask us

Also coming: tajweed correction ($1,000 planned). Get notified.

Private means private

The recitation stays with the reciter.

On the device.

The model runs in the browser tab, on the phone's own CPU. There is no speech server, so no audio is uploaded, and there is no per-minute bill.

Offline once loaded.

After the first download the model needs no network. The mushaf pages and translation are cached the same way.

Open weights.

Read the model card, run your own tests, keep your own copy. Nothing phones home.

Your mistakes are yours.

The tajweed Mistake History lives in the browser's storage on that device. Clear it and it is gone.

Questions

Asked before buying.

What exactly is free?

The three Naml models: weights, the WebAssembly core that runs them, the unit lists and a model card, under CC BY 4.0 (credit Fanah Solutions and NVIDIA's NeMo FastConformer, which the models derive from). The tracker code and the masjid kit are paid; tajweed correction is coming.

Which phones does it run on?

Any browser with WebAssembly SIMD: current Chrome, Edge, Firefox and Safari. Measured per 0.32 s step: about 18 ms on a budget Android phone (4 GB, Chrome), about 15 ms on an iPhone in Safari, about 15 ms on one server CPU core. Older browsers without SIMD fall back to a slower path.

How accurate is the timing?

On 31 reviewed clips (about 1,240 ayahs), Naml Letters puts both the start and the end of an ayah within 250 ms for 51% of ayahs and within 500 ms for 68%. wav2vec2 with Viterbi alignment: 43% and 62%.

What do I get for $500?

A download of the recite app's source (TypeScript, React, the Rust source of the wasm core), its README and a licence to use it in your products. Not included: test recordings, our development plugins, third-party models. See the licence draft.

What is in the masjid kit?

The recite app's source set up as a Tarawih screen: the Prayer view (it listens from Al-Fatiha, follows the imam into the passage and shows each ayah with its English) and the large single-ayah view, plus a setup guide for one laptop, a TV or projector and an audio feed from the mixing desk. It runs offline. If you would rather not set it up yourself, choose "Built for you".

What will tajweed correction detect, and what not?

On by default: shadda, ikhfa, iqlab, idgham, confused letters (ص/س, ط/ت, ح/ه, ع/ء), short vowels, a dropped letter or vowel, and missing, incorrect or extra words. Not detected yet: how long a madd or a ghunna is held. On correct professional recitation it raises about one false alarm every 5 to 6 minutes.

Can I get a refund?

See the refund policy draft. If the code does not run as described, write to us.

Do you host it for me?

No hosted API: everything runs on the user's device. If you want it built into your app or a masjid screen, choose "Built for you".

From the blog

How it works.

All posts →

Built for you

Ask me directly.

Tell me what you are making: an app, a hifz class, a masjid screen. I will say whether the code already does it, or quote the work.

Or write to hello@fanahsolutions.com.