urdunorm

Urdu text normalization and WER/CER scoring for ASR evaluation, under named protocols you can report. Zero runtime dependencies. pip install urdunorm

The problem, in five lines

from urdunorm import wer

reference = ["کے لیے شکریہ"]  # what the transcriber wrote
hypothesis = ["کیلئے شکریہ"]  # what the model emitted, compound joined

wer(reference, hypothesis)  # 0.0
wer(reference, hypothesis, mode="lums-compat")  # 0.667

Same audio, same words, one spelling difference, two published protocols. That gap is why Urdu ASR results do not compare across papers, and why this package makes you name the protocol.

What it does to text, under eval

input	output	what happened
علي كي	علی کی	Arabic-keyboard letters unified to their Urdu equivalents. Same word, different keyboard.
ﻤﯿﮟ	میں	Presentation forms, common in PDF-extracted text, folded to normal letters.
مُحَمَّد	محمد	Vowel diacritics stripped. Optional in writing, so a model is not charged for omitting them.
۲۰۲۶	2026	All three numeral systems Urdu uses folded onto one.
کیلئے	کے لیے	A compound written without its space, repaired. WER counts whitespace tokens, so this alone costs two errors.
بھائی	بھائی	Unchanged. That letter marks aspiration and is phonemic. Merging it into its lookalike reports a lower WER by destroying the distinction.

The last row matters most: phonemic characters are a contract, and the package refuses to import if a rule table would merge one.

How it differs from urduhack and LughaatNLP

Those are general Urdu NLP toolkits: tokenizers, POS tagging, NER. This is an evaluation protocol, normalization plus scoring, nothing else. Different jobs.

Differences are measured, not asserted. This table is generated by compare_normalizers.py --emit-table running each package at a pinned version:

what it is	input	urdunorm	urduhack	LughaatNLP
Arabic-keyboard yeh and kaf	علي كي دكان	علی کی دکان	علی کی دکان	علي كي دكان
presentation forms	ﻤﻴﮟ	میں	مےں	ﻤﻴﮟ
vowel diacritics	مُحَمَّد	محمد	محمّد	محمد
Urdu-Indic digits	۲۰۲۶	2026	۲۰۲۶	۲۰۲۶
joined compound	کیلئے	کے لیے	کیلئے	کیلئے
aspiration, must not change	بھائی	بھائی	بھائی	بھائی
alef madda, must not change	آج	آج	آج	اج

Measured on urdunorm 0.1.0, urduhack 1.1.1, LughaatNLP 1.3.1.

Both also carry heavy dependencies: urduhack imports TensorFlow at package import; LughaatNLP requires torch, transformers, scikit-learn and scipy.

Modes

Mode	Characters	Diacritics	Punctuation	Digits	Space repair	ZWNJ
eval	unified	stripped	dropped	unified	on	to space
display	unified	kept	kept	kept	off	kept
lums-compat	unified	stripped	5 chars only	unified	off	to space

Use eval unless you know why not. display is for showing transcripts to people. lums-compat reproduces WER We Stand (COLING 2025), the only published Urdu ASR benchmark, so numbers can be compared against it. Its five-character punctuation class keeps the Urdu full stop: that is theirs, not a bug here. Report under two protocols; a single-protocol Urdu WER compares to nothing.

Scoring

from urdunorm import wer, cer, score, normalize, EVAL

normalize("کیلئے", mode="eval")
wer(references, hypotheses, mode="eval")
cer(references, hypotheses, mode="eval")

result = score(references, hypotheses, mode="eval")
str(result)  # 'WER 12.40% [eval]'
EVAL.with_(repair_spaces=False)  # an ablation is a field override

Both sides are normalized under the named mode and the result records which protocol produced it. Tokenization is a whitespace split. The corpus rate is total errors over total reference tokens, never the mean of per-utterance rates, which would weight a three-word clip like a thirty-word one.

More

Why this exists · Measured comparison · Where the rules come from · Review the Urdu rules (no Python needed) · Contributing

The space-repair lexicon and the number tables have not been signed off by a native speaker, so numbers_to_words ships off in every mode.

Citing

@software{urdunorm,
  title  = {urdunorm: Urdu text normalization for reproducible ASR evaluation},
  author = {Hassan Bhatti, Hunzalah},
  year   = {2026},
  doi    = {10.5281/zenodo.21864104},
  url    = {https://github.com/hunzed/urdunorm}
}

Apache-2.0.