Sponsored by
iMean AI Trip Planner
- AI trip planner that creates personalized itineraries, finds cheap flights
Free Tools
Products
Category
- New AIs
The Latest AIs, every day
- Most Saved AIs
AIs with the most favorites on Toolify
- Most Used AIs
AIs with the highest website traffic (monthly visits)
- AI Apps
AI Tools by Apps
- Discord of AI
Discover the Discord of AI
- AI Chrome Extensions
AI Tools by browser extensions
- GPTs
GPTs from GPT Store
- AI Models
Discover The Best Model For AI
Ranking
Jobs
- Top AI By Monthly
Top AI lists by month and monthly visits.
- Top AI By Categories
Top AI lists by category and monthly visits.
- Top AI By Regions
Top AI lists by region and monthly visits.
- Top AI By Source
Top AI lists by source and monthly visits.
- Top AI by Revenue
Top AI lists by revenue and real traffic.
AI Ad Library
AI Models
zipformer_p-arabic-v3
Quran-Lab / zipformer_p-arabic-v3
Model's Last Updated: August 28 2026
automatic-speech-recognition
Introduction of zipformer_p-arabic-v3
Model Details of zipformer_p-arabic-v3
zipformer_p-arabic-v3
A 65.5M-parameter streaming phoneme recogniser for Qur'an recitation (Hafs 'an 'Asim), built on Zipformer2 with a CTC head. It transcribes recitation directly into a 251-symbol Qur'anic phonetic alphabet that encodes tajweed-relevant distinctions: madd length, gemination, ghunna, ikhfaa, qalqalah, and the emphatic consonants.
Design principle: no language model, anywhere. No transducer, no LM rescoring, no text prior. An internal or external LM silently corrects a reciter's mistakes, which is precisely what a recitation-assessment system must never do. Accuracy comes from data and the encoder alone.
Results
Phoneme Error Rate on the open Quranic ASR Benchmark (v1.1 references), against the previous public release of this line:
| held-out set | Quran Model | zipformer_p-arabic-v2 |
|---|---|---|
| studio reciters, 3 benchmark reciters* | 1.43% | 5.19% |
| real phone recordings | 3.65% | 7.92% |
| unseen professional reciter | 9.10% | 11.91% |
Measured on the shipped checkpoint (SHA-verified) with the public per-eval manifest; an earlier revision of this card carried 4.92% for the unseen reciter, which does not reproduce on the shipped weights and has been corrected.
* The three studio benchmark reciters were intended to be fully absent from training; a post-release audit found 24 stray clips of them (0.001% of the 1.85M-row finetune manifest, none of the 600 benchmark clips) introduced by the concatenation builder. Reported as-is for transparency.
The emphatic-consonant gap (ص ض ط ظ error rate minus their plain counterparts), a known weakness of earlier models, is statistically indistinguishable from zero on all three sets.
On obadx/qdat_bench (159 everyday reciters, one ayah, human-annotated tajweed attributes), madd-length measurement by attribute (RMSE in harakat, lower is better):
| attribute | Quran Model | muaalem-model-v3_2 |
|---|---|---|
| qalo_alif | 0.210 | 0.449 |
| qalo_waw | 0.514 | 0.456 |
| allam_alif | 0.363 | n/a |
| separate madd (munfasil) | 0.887 | 0.687 |
| madd aared | 1.458 | n/a |
Honest limitations
These are measured, not hypothetical. Read them before building on the model.
- It is trained almost entirely on correct recitation with canonical labels. When a reciter deviates from the rule, the model tends to transcribe the rule rather than the deviation. On qdat_bench it reported full ikhfaa for 81 of 94 clips where human annotators heard a plain noon. Do not use it, as shipped, as a standalone judge of whether ikhfaa or similar quality-contrasts were performed.
- Free-choice madd lengths (munfasil, aared) are unreliable in the token stream because the training labels fix them at 4 harakat while reciters legitimately choose 2, 4 or 6. Duration grading should be done from forced-alignment timing, not from the emitted symbol run-length.
- Children recite into its weakest region : under-12 reciters score 2 to 3 times worse than adults on qdat_bench.
- Chunk size matters and depends on domain. On in-domain recitation, streaming chunks (16 or 24 frames) match or beat full context. On out-of-domain audio, full context is about 2 PER points better. The model was trained only at chunk sizes 8, 16 and 24.
Files
| file | purpose |
|---|---|
zipformer_p_arabic_v3.pt
|
PyTorch weights (average of the last 3 training epochs) |
zipformer_p_arabic_v3.onnx
/
zipformer_p_arabic_v3.int8.onnx
|
cache-aware streaming CTC export, sherpa-onnx compatible |
tokens.txt
|
symbol table.
Use it.
The raw
phoneme_units.json
ids are offset by one against the CTC output layer (blank is 250, not 0); decoding without this table produces garbage
|
phoneme_units.json
|
tokenizer unit inventory |
ordered_quran_phonemes.json
|
canonical phonemisation of all 6,236 ayat (for retrieval / grading) |
quran_text2phoneme.json
|
text-to-phoneme lookup used in evaluation |
quran_per_eval.py
,
quran_wer_retrieval.py
|
evaluation scripts |
decode_with_confidence.py
|
CTC decoding with per-symbol confidence |
export_quran_streaming_onnx.py
|
the exact export script that produced the ONNX files |
Usage (ONNX, streaming)
The ONNX models are standard sherpa-onnx streaming-CTC zipformer2 exports (
model_type=zipformer2
,
chunk 48 input frames). Feed 16 kHz mono, 80-dim kaldi fbank (povey window). The features must be
kaldi fbank, not a mel-spectrogram: a slaney-mel front end roughly doubles the error rate.
import onnxruntime as ort, numpy as np
sess = ort.InferenceSession("zipformer_p_arabic_v3.int8.onnx")
# see quran_per_eval.py for the full streaming loop incl. cache tensors and fbank settings
Training summary
Trained on about 5,400 effective hours per epoch for 10 epochs: professional complete-mushaf recitations (36 EveryAyah reciters, 1,161 additional reciters segmented from public recitation archives), real phone recordings, concatenated multi-ayah context windows, same-ayah repetition clips, and a broad-Arabic tier, with noise, reverberation, wind, speed and tempo augmentation. Labels are the deterministic Qur'anic phonetic script of quran-transcript (Hafs, murattal).
License
Released under the Quran-Lab No-Profit License, Version 1.1 (NPL-1.1) ; see LICENSE . In short: free to use, study, adapt and redistribute, for any purpose that does not generate profit.
Related
- Benchmark and scorer: Quran-Lab/quranic-asr-benchmark
- Leaderboard: Quran-Lab/quranic-asr-leaderboard
- Previous releases in this line: zipformer_p-arabic-v2 , zipformer_p-quran
- The qdat_bench tajweed benchmark and the muaalem models by obadx , whose phonetic script this model's labels are built on
Runs of Quran-Lab zipformer_p-arabic-v3 on huggingface.co
346
Total runs
4
24-hour runs
20
3-day runs
114
7-day runs
334
30-day runs
More Information About zipformer_p-arabic-v3 huggingface.co Model
More zipformer_p-arabic-v3 license Visit here:
https://choosealicense.com/licenses/quran-lab-npl-1.2zipformer_p-arabic-v3 huggingface.co
zipformer_p-arabic-v3 huggingface.co is an AI model on huggingface.co that provides zipformer_p-arabic-v3's model effect (), which can be used instantly with this Quran-Lab zipformer_p-arabic-v3 model. huggingface.co supports a free trial of the zipformer_p-arabic-v3 model, and also provides paid use of the zipformer_p-arabic-v3. Support call zipformer_p-arabic-v3 model through api, including Node.js, Python, http.
zipformer_p-arabic-v3 huggingface.co Url
https://huggingface.co/Quran-Lab/zipformer_p-arabic-v3Quran-Lab zipformer_p-arabic-v3 online free
zipformer_p-arabic-v3 huggingface.co is an online trial and call api platform, which integrates zipformer_p-arabic-v3's modeling effects, including api services, and provides a free online trial of zipformer_p-arabic-v3, you can try zipformer_p-arabic-v3 online for free by clicking the link below.
Quran-Lab zipformer_p-arabic-v3 online free url in huggingface.co:
https://huggingface.co/Quran-Lab/zipformer_p-arabic-v3zipformer_p-arabic-v3 install
zipformer_p-arabic-v3 is an open source model from GitHub that offers a free installation service, and any user can find zipformer_p-arabic-v3 on GitHub to install. At the same time, huggingface.co provides the effect of zipformer_p-arabic-v3 install, users can directly use zipformer_p-arabic-v3 installed effect in huggingface.co for debugging and trial. It also supports api for free installation.
zipformer_p-arabic-v3 install url in huggingface.co:
https://huggingface.co/Quran-Lab/zipformer_p-arabic-v3Url of zipformer_p-arabic-v3
zipformer_p-arabic-v3 huggingface.co Url
Provider of zipformer_p-arabic-v3 huggingface.co
Other API from Quran-Lab
Credits Limited