Bayanat Labs Arabic Data Cost and Pilot Estimator
Turn your assumptions about an Arabic annotation or transcription project into line-item hours, an illustrative cost range, a week-by-week timeline and a pilot statement of work with an acceptance checklist. Line items, costs, phases, weeks and pilot totals show their formulas, and the range and the acceptance plan state their methods. Every output is an illustrative estimate from your assumptions, not a quote; every rate is yours to enter, and no Bayanat Labs price appears. 40 planning scenarios and 27 invalid inputs test the JavaScript engine and the Python port against a separately written calculation. The app runs in your browser, loads nothing from other hosts, uses no cookies, browser storage or analytics, and never puts your inputs in the URL.
| Licence | Apache-2.0 (code and app); CC BY 4.0 (fixtures, examples and page text); SIL OFL (fonts) |
|---|---|
| Version | v0.1.0 · Code, fixtures and app v0.1.0, tagged on GitHub; the Python port installs from GitHub and is not on PyPI. |
| Published | |
| Updated | |
| Who it is for | Teams planning an Arabic annotation or transcription project who want line-item hours, a range, a timeline and a pilot scope from their own assumptions, with every formula shown. |
| Level | Beginner |
Key numbers
Counted by script from the released files. Each key number is read or computed from fixtures released at tag v0.1.0, commit 0db3d41836b85707bc8410a3491b8282e6876583, and pinned to its SHA-256; hours, costs and shares are rounded half up as the README writes them. fixtures/SHA256SUMS lists a checksum for every fixture. Space revision b9c86b28469fdd737ad4d21a789a2152b8f508a6.
Downloads and identifiers
| Where | What |
|---|---|
| Hugging Face | Open the estimator |
| GitHub | Code on GitHub |
| Hugging Face | Open the estimator in Arabic |
| GitHub | Method |
| GitHub | Fixtures and checksums |
| bayanatlabs.com | Arabic Data Annotation Cost, the article behind the worked example |
Get started
The app needs no install. Install the Python port from GitHub at the v0.1.0 tag; the package is not on PyPI. Run the command from a clone of the repository, which holds the example.
pip install "bayanatlabs-estimator @ git+https://github.com/Bayanat-Labs/arabic-data-cost-estimator@v0.1.0"
bl-estimate examples/chat_messages.json --format md --lang enWhat is released
| Part | Licence | In this release |
|---|---|---|
| Browser app on Hugging Face Spaces, interface in English and Arabic: plain JavaScript with no build step | Apache-2.0 | Yes |
| Python port and command-line tool (bl-estimate), with no dependencies | Apache-2.0 | Yes |
| Separately written calculation, schemas, tests and method document | Apache-2.0 | Yes |
| Fixtures and examples: planning scenarios and rejected inputs, invented, with no personal data | CC BY 4.0 | Yes |
| Text of this record and of the app's page | CC BY 4.0 | Yes |
| IBM Plex Sans Arabic font in the app (SIL Open Font License 1.1) | SIL Open Font License, copyright IBM Corp. | Yes |
Attribution: Arabic Data Cost and Pilot Estimator by Bayanat Labs. https://github.com/Bayanat-Labs/arabic-data-cost-estimator
Intended uses
- Plan the hours, range and timeline of an Arabic annotation or transcription project from your own assumptions.
- Draft a pilot statement of work with an acceptance plan and checklist, every assumption listed.
- Run the same estimate in scripts with the Python port, which writes JSON, CSV or Markdown.
Out of scope
- Reading an output as a quote, a price or a Bayanat Labs commitment: every output is an illustrative estimate from your assumptions.
- Treating the range or the acceptance plan as a statistical guarantee: both are planning approximations.
Rights at a glance
Training, display, redistribution and synthetic-voice use are separate permissions.
| Use | Status | Note |
|---|---|---|
| Training | Not applicable | A planning tool and invented fixtures, not training data. The fixtures are CC BY 4.0 if you reuse them. |
| Display | Allowed | Fixtures, examples and page text under CC BY 4.0, with attribution. |
| Redistribution | Allowed | Code under Apache-2.0; fixtures, examples and page text under CC BY 4.0; the bundled font under the SIL Open Font License. The Bayanat Labs name and logo are not licensed under any of these. |
| Synthetic voice | Not applicable | No audio. |
Limitations
- Illustrative estimate from your assumptions, not a quote. Every output is an illustrative estimate from your assumptions. Replace the defaults with what you know of your own project.
- Every rate is entered by you. The tool ships no price list, band or vendor figure, and no Bayanat Labs price appears in it. The placeholder rate in the app is the article's hypothetical rate, labelled as illustrative; it is not a Bayanat Labs rate or a market benchmark. The rates in the fixtures are placeholders chosen to exercise the arithmetic.
- QA seconds in the speech presets are illustrative placeholders: adjudication, senior QA and rework are each set to one listen-through of the audio.
- Audit hours are blank for jobs billed in audio or per thousand words. They are computed only when the billing unit is itself the unit sampled, an item or an utterance; the audit hours are shown beside the plan and not added to the total.
- The range is a seeded Monte Carlo with triangular ranges, and the same seed gives the same range. It is illustrative, not a statistical guarantee.
- The acceptance plan uses an exact binomial sample for accuracy or error rate and normal approximations for kappa, WER and CER. These are planning approximations.
- In the worked example, three inputs are inferred from the article's figure rather than stated, and the longer guidelines are read as doubled, because the article says only "longer". The method document lists them.
- The Arabic interface and Arabic exports have not yet been reviewed by a native Arabic writer; the English governs.
Sources and licences
| Source | Licence | Redistributed here |
|---|---|---|
| Arabic Data Annotation Cost, Bayanat Labs article, read on 4 November 2026 | The worked example reproduces its illustrative budget and hypothetical rates; not included | No |
The defaults repeat the assumptions of the worked example, so the page opens on a result you can check against the article; they are starting points to replace. The speech effort presets are the careful and quick figures for conversational telephone speech in Cieri et al., as cited in the article. Source details
How to cite
Cite as:
Bayanat Labs (2026). Arabic Data Cost and Pilot Estimator, version 0.1.0. Bayanat Labs. https://github.com/Bayanat-Labs/arabic-data-cost-estimator
No DOI. Cite the URL in the citation above and the date you read it.
@misc{bayanatlabs2026arabicdatacostestimator,
author = {{Bayanat Labs}},
title = {Arabic Data Cost and Pilot Estimator, version 0.1.0},
year = {2026},
month = nov,
howpublished = {\url{https://github.com/Bayanat-Labs/arabic-data-cost-estimator}}
}
Changelog
| Date | Version | Change |
|---|---|---|
| v0.1.0 | Browser app in English and Arabic on Hugging Face Spaces; Python port, command-line tool, fixtures and method on GitHub. |
Corrections and removal requests
If something in this release should not be there, or is wrong, email hello@bayanatlabs.com. We handle removal requests within 30 days, and platform requests within 24 hours. How removal requests work
Related product: Bayanat Knowledge