---
title: "Arabic speech-to-text API - how to choose in 2026 | BroadRouter"
description: "How to pick an Arabic transcription API: word error rate, dialects, file limits, price per audio hour and privacy, plus a test plan for your own recordings."
url: https://broadrouter.com/blog/arabic-speech-to-text-api-buyers-guide/
lang: en
updated: 2026-11-14
publisher: BroadRouter (a BroadNet site); API by BroadNet Inference, https://inference.broadnet.ai
---

# Arabic speech-to-text API: how to choose one in 2026

By the BroadNet Inference team · 7 min read · Published 2026-11-14 · Updated 2026-11-15

Pick an Arabic speech-to-text API by testing it on your own recordings: measure word error rate on your dialects, then compare price per audio hour, file limits and data retention. BroadNet Transcribe Arabic costs $0.25 per audio hour ($0.0042 a minute), and the Open Universal Arabic ASR Leaderboard listed it #4 of 39 on 15 November 2026.

## Key takeaways

-   Word error rate (WER) on your own audio matters more than any leaderboard. Lower is better.
-   Modern Standard Arabic is the easy case. Test Gulf, Egyptian, Levantine and Maghrebi speech separately.
-   Normalise Arabic text before scoring, or spelling variants will count as errors.
-   BroadNet Transcribe Arabic costs $0.25 per audio hour; 1,000 hours cost $252.00.
-   Always send `language=ar`, or Arabic may come back transliterated into Latin letters.
-   The free $5 starting credit covers about 1,190 minutes of Arabic audio for testing.

**On this page**

1.  Why does word error rate on your own audio matter most?
2.  How do you check dialect coverage?
3.  What does Arabic transcription cost per audio hour?
4.  Which features should an Arabic speech-to-text API have?
5.  What file limits and formats apply?
6.  Is the audio stored?
7.  How do you test an Arabic transcription API in one afternoon?
8.  Common mistakes
9.  Where BroadNet Inference fits

To choose an Arabic [speech-to-text](https://broadrouter.com/glossary/speech-to-text/) API, transcribe 30 to 50 of your own clips per dialect with each candidate, compare word error rate per dialect, then weigh price per audio hour, file limits, streaming needs and data retention. A model that looks great on a benchmark can still fail on your call-centre audio.

Arabic is hard for speech recognition: one written standard sits on top of many spoken dialects, speakers switch to English mid-sentence, and most real audio is phone calls and voice notes.

## Why does word error rate on your own audio matter most?

[Word error rate (WER)](https://broadrouter.com/glossary/word-error-rate/) is the share of words the model gets wrong: substituted, missed or invented. Lower is better. Public numbers are a starting point:

| Test set | What it is | BroadNet Transcribe Arabic | ElevenLabs Scribe v2 | Whisper large-v3 |
| --- | --- | --- | --- | --- |
| FLEURS-ar | Clean read sentences | 3.8 | not published | not published |
| MASC-ar | Mixed real-world Arabic speech | 21.1 | 22.3 | 31.5 |
| BroadNet dialect set | Gulf, Egyptian and Levantine speech | 23.4 | 25.6 | 31.8 |

These are BroadNet’s lab measurements, published on its [model catalogue](https://www.broadnet.ai/inference/models) and its [Transcribe Arabic page](https://www.broadnet.ai/inference/models/cohere-transcribe-arabic). BroadNet ran the comparison itself, so treat the ElevenLabs and Whisper columns as one vendor’s test of its competitors, not an independent result. BroadNet also reports the model at #2 of 37 (its September 2026 figure; the live board listed it #4 of 39 on 15 November 2026) on the [Open Universal Arabic ASR Leaderboard](https://huggingface.co/spaces/elmresearchcenter/open_universal_arabic_asr_leaderboard), and #1 among models whose licences permit serving through an API. On 15 November 2026 the three entries above it were Jais-ASR, Sygma-ASR-Arabic (which the board describes as beam decoding on the Cohere model) and Audar-ASR-V1-Turbo; Whisper large-v3 was #19. We have not checked each entry’s licence.

You can hear the model on your own voice before signing up: BroadNet’s [browser playground](https://www.broadnet.ai/inference/playground) transcribes up to 60 seconds of Arabic or English audio with no account. See [how to use the playground](https://broadrouter.com/guides/try-broadnet-inference-playground/).

The gap between 3.8 and 23.4 is the lesson: **the same model can be near-perfect on read speech and much weaker on dialect conversation.** Test on yours.

Be careful with vendor numbers in general. Many published Arabic WER figures come from the vendor’s own test, on audio you can’t see. A number is only comparable when the test set, the normalisation and the scoring tool are the same.

For the full picture, our open [speech-to-text API price index](https://broadrouter.com/data/speech-to-text-api-prices/) lists batch and streaming prices, billing increments and Arabic support for more than 20 products, each checked against the provider’s own pricing page (CSV and JSON).

## How do you check dialect coverage?

Group your test files by dialect: Gulf (Khaleeji), Egyptian, Levantine, Iraqi, Maghrebi, plus [Modern Standard Arabic](https://broadrouter.com/glossary/msa/). Score each group on its own. If most of your customers are in Riyadh, Gulf WER matters far more than an average.

Watch for **[code-switching](https://broadrouter.com/glossary/code-switching/)**. Gulf business speech mixes in English product names and numbers. Check how numbers, dates and names come out, because those are the words your downstream system acts on. BroadNet offers a [fine-tuned Arabic model](https://broadrouter.com/models/#on-request) for numbers, dates and names in noisy audio, on request.

## What does Arabic transcription cost per audio hour?

Compare per hour of audio, because that is how your volume is measured:

| Model | Per minute | Per hour | 1,000 hours |
| --- | --- | --- | --- |
| [BroadNet Transcribe Arabic](https://broadrouter.com/models/cohere-transcribe-arabic/) | $0.0042 | $0.25 | $252.00 |
| [BroadNet Transcribe (14 languages)](https://broadrouter.com/models/cohere-transcribe/) | $0.002 | $0.12 | — |
| Soniox async (file), 60+ languages | about USD 0.0017 (derived) | about USD 0.10 | about USD 100 |
| Soniox real-time (streaming) | about USD 0.002 (derived) | about USD 0.12 | about USD 120 |

Soniox rows are from [Soniox pricing](https://soniox.com/pricing), checked 14 November 2026. Soniox bills per token (about 30,000 audio tokens per hour) and quotes the hourly figures as approximations.

**Soniox is cheaper per hour than BroadNet.** Its file rate is less than half of BroadNet Transcribe Arabic’s $0.25 and below BroadNet Transcribe’s $0.12. Its real-time rate matches BroadNet Transcribe and still undercuts the Arabic model, and BroadNet does not document real-time streaming at all. Soniox lists Arabic (`ar`) among its [supported languages](https://docs.soniox.com/stt/concepts/supported-languages) and shows one rate for all languages. It does not publish Arabic-specific pricing, Arabic WER figures or dialect coverage, so the price is clear but the accuracy is not. Run the same dialect test on both before you choose on price.

For a support inbox, 10,000 WhatsApp voice notes of 30 seconds each cost $21.00 on the Arabic model. The math is `10,000 × 0.5 minutes × $0.0042`. Run your own volume through the [speech-to-text cost calculator](https://broadrouter.com/tools/speech-to-text-cost-calculator/).

When comparing providers, check the billing unit as well as the rate: per second, per minute or per started block, and whether there is a minimum charge per file. Short voice notes are where billing rounding shows up. For other providers’ per-minute rates, checked against their pricing pages, see [speech-to-text API pricing compared](https://broadrouter.com/blog/speech-to-text-api-pricing-compared/); if you are moving off OpenAI’s endpoint, see the [Whisper API alternative](https://broadrouter.com/blog/whisper-api-alternative/) checklist. The [Transcribe vs Transcribe Arabic comparison](https://broadrouter.com/compare/cohere-transcribe-vs-cohere-transcribe-arabic/) shows when the cheaper 14-language model is enough.

## Which features should an Arabic speech-to-text API have?

Not every product needs every feature. Decide which ones matter before you compare:

| Feature | Why it matters | What to ask |
| --- | --- | --- |
| Batch file upload | Voice notes, recordings, archives | Formats and maximum file size |
| Real-time streaming | Live captions, voice agents | Latency, and whether it’s a separate product |
| Speaker labels ([diarization](https://broadrouter.com/glossary/diarization/)) | Calls with agent and customer | Batch only, or live too? |
| Word timestamps | Subtitles, search, review tools | Word or segment level? |
| Code-switching | Arabic with English terms | Does it need one language per request? |
| Data retention | Calls carry personal data | Is audio or text kept, and for how long? |

What BroadNet documents today: batch transcription through the OpenAI-compatible `/v1/audio/transcriptions` endpoint, WAV or MP3 input, a `language` field, and zero data retention. For anything else on this list, check the [API reference](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-arabic-speech-to-text-api-buyers-guide&utm_content=body) before you rely on it.

## What file limits and formats apply?

-   **File size.** BroadNet accepts WAV or MP3 up to 26,214,400 bytes (25 MiB) per file. A larger file returns 413; split longer recordings.
-   **Formats.** Convert .ogg voice notes and .m4a recordings first; the [WhatsApp voice note guide](https://broadrouter.com/guides/transcribe-whatsapp-voice-notes/) has the ffmpeg line.
-   **The language field.** Send `language=ar`. Without it, Arabic can come back transliterated into Latin letters.
-   **Rate limits.** A pay-as-you-go key allows 30 requests a minute and 4 at once. For a large backlog, queue files and retry 429s with backoff; see [errors and retries](https://broadrouter.com/guides/errors-and-retries/).

## Is the audio stored?

Voice notes and calls carry names, phone numbers and health or banking details. Ask every provider whether audio and transcripts are retained, and for how long. BroadNet Inference does not retain either; see [zero data retention](https://broadrouter.com/zero-data-retention/) and our [ZDR checklist](https://broadrouter.com/blog/zero-data-retention-llm-api-checklist/).

## How do you test an Arabic transcription API in one afternoon?

1.  Collect 30 to 50 real clips per dialect from your production channel. Remove anything you may not share.
2.  Write the correct transcript for 10 clips per dialect by hand, using one agreed spelling convention.
3.  Transcribe every clip with each candidate API. The [step-by-step Arabic guide](https://broadrouter.com/guides/arabic-speech-to-text-api/) has curl and Python code for BroadNet.
4.  Normalise both texts the same way and compute WER and CER per dialect. [Arabic WER explained](https://broadrouter.com/blog/arabic-asr-wer-explained/) has the Python normalisation function and shows why skipping it can turn a perfect transcript into 75% WER.
5.  Read 20 transcripts yourself. A low WER can still hide wrong numbers or names.

How to pick clips, write references and score each dialect group is covered in detail in [Arabic dialect speech recognition](https://broadrouter.com/blog/arabic-dialect-speech-recognition/).

## Common mistakes

-   **Testing only clean audio.** Use real phone and voice-note recordings.
-   **One average WER.** A good average can hide a dialect your customers use.
-   **Skipping the language field.** Transliterated output scores as almost all errors.
-   **Ignoring numbers.** Check digits, dates and amounts by hand; they are what your system acts on.

## Where BroadNet Inference fits

BroadNet Transcribe Arabic suits batch transcription of Arabic voice notes, calls and recordings at a per-minute price, with no audio retained. The $5 starting credit covers about 1,190 minutes of Arabic audio, enough for the test above without a card. If you need features it doesn’t document, such as live streaming captions, test a provider that offers them. For Arabic text tasks after transcription, see [choosing an LLM for Arabic](https://broadrouter.com/blog/choosing-an-llm-for-arabic/).

Free tools for this

-   [Arabic token counter](https://broadrouter.com/tools/arabic-token-counter/) Arabic vs English token counts per tokenizer: the "Arabic tax" in tokens and money.
-   [Arabic text normalizer](https://broadrouter.com/tools/arabic-text-normalizer/) Tashkeel, tatweel, alef, ya, ta marbuta, digits and invisible characters, with a matching Python function.
-   [Speech-to-text cost calculator](https://broadrouter.com/tools/speech-to-text-cost-calculator/) Hours, calls or voice notes to a monthly transcription bill, against your current provider.

[All free tools](https://broadrouter.com/tools/)

## Terms in this article

[Speech-to-text](https://broadrouter.com/glossary/speech-to-text/)

Speech-to-text, also called automatic speech recognition (ASR), is technology that converts spoken audio into written text. A speech-to-text API accepts an audio file or stream and returns a transcript, usually priced per minute of audio. Accuracy is measured with word error rate and varies with accent, dialect and recording quality.

[Word error rate](https://broadrouter.com/glossary/word-error-rate/)

Word error rate (WER) is the standard accuracy measure for speech recognition: the number of word substitutions, deletions and insertions needed to turn a transcript into the correct reference, divided by the number of words in the reference. Lower is better.

[Streaming](https://broadrouter.com/glossary/streaming/)

Streaming is a way of returning a language model response in small chunks while it is still being generated, instead of as one reply at the end. OpenAI-compatible APIs stream over server-sent events when the request sets stream to true.

[Code-switching](https://broadrouter.com/glossary/code-switching/)

Code-switching is the use of two or more languages, or varieties of one language, within the same conversation or sentence, in speech or writing. Examples include Arabic speakers mixing English words into dialect, or moving between Modern Standard Arabic and their dialect.

[Pay-as-you-go](https://broadrouter.com/glossary/pay-as-you-go/)

Pay-as-you-go is a pricing model in which you pay only for what you use, metered per unit such as tokens, requests or audio minutes, with no subscription or minimum commitment. Most LLM APIs bill this way, from a prepaid balance or a periodic invoice.

[Token](https://broadrouter.com/glossary/token/)

A token is the unit of text a language model reads and writes: a word, part of a word, a punctuation mark or a byte sequence produced by the model's tokenizer. LLM APIs count, limit and usually bill usage in tokens.

[Zero data retention](https://broadrouter.com/glossary/zero-data-retention/)

Zero data retention (ZDR) is a data-handling policy in which an AI provider processes prompts and outputs only to serve a request and does not store that content afterward, in logs, caches or training data. Content-free records such as token counts and timings are usually still kept.

[Dialectal Arabic](https://broadrouter.com/glossary/dialectal-arabic/)

Dialectal Arabic is the everyday spoken Arabic of a region or city, such as Egyptian, Gulf, Levantine or Maghrebi Arabic. Dialects differ from Modern Standard Arabic in sounds, words and grammar, have no standard spelling, and are increasingly written online, which makes them harder for language models.

## Questions

### What is a good word error rate for Arabic?

It depends on the audio. BroadNet reports 3.8 on FLEURS-ar (clean read speech), 21.1 on MASC-ar and 23.4 on its dialect set for BroadNet Transcribe Arabic. Conversational dialect audio is always harder than read speech.

### Which speech-to-text API is best for Arabic dialects?

There is no single answer, because results vary by dialect and recording quality. Shortlist two or three APIs, transcribe 30 to 50 of your own clips per dialect with each, and compare word error rate per dialect, not an overall average.

### How much does it cost to transcribe 1,000 hours of Arabic audio?

At $0.0042 per audio minute, 1,000 hours cost $252.00 with BroadNet Transcribe Arabic.

### Does BroadNet's Arabic speech-to-text support live streaming?

BroadNet documents batch transcription through /v1/audio/transcriptions with WAV or MP3 files up to 25 MiB. It does not document live streaming captions. If you need them, Soniox, for example, sells real-time transcription with Arabic in its language list (checked 14 November 2026).

### Can it transcribe WhatsApp voice notes?

Yes, after converting the .ogg file to WAV or MP3 with ffmpeg. The WhatsApp guide on this site shows the one-line conversion.

### Is my audio stored?

No. BroadNet Inference does not retain audio or transcripts after the request; it keeps only content-free records such as timings and cost.

## Keep reading

-   [Transcribe long audio over 25 MB: split, send and stitch](https://broadrouter.com/blog/transcribe-long-audio-files-api/)
-   [Arabic ASR word error rate (WER) explained](https://broadrouter.com/blog/arabic-asr-wer-explained/)
-   [Arabic dialect speech recognition: Gulf, Egyptian, Levantine, Maghrebi](https://broadrouter.com/blog/arabic-dialect-speech-recognition/)
-   [Speech-to-text API pricing compared: price per minute in 2026](https://broadrouter.com/blog/speech-to-text-api-pricing-compared/)
-   [Whisper API alternative: when to switch, and how to test one](https://broadrouter.com/blog/whisper-api-alternative/)
-   [Choosing an LLM API for Arabic: what to test](https://broadrouter.com/blog/choosing-an-llm-for-arabic/)
-   [Add diacritics to Arabic text with an LLM: prompt, checks, DER](https://broadrouter.com/blog/arabic-diacritization-llm/)
-   [Arabic grammar and spelling correction with an LLM: a how-to](https://broadrouter.com/blog/arabic-grammar-spelling-correction-llm/)
-   [Guide: Arabic speech-to-text API, step by step](https://broadrouter.com/guides/arabic-speech-to-text-api/)
-   [Guide: Transcribe WhatsApp voice notes to text with an API](https://broadrouter.com/guides/transcribe-whatsapp-voice-notes/)
-   [Guide: BroadNet Inference error codes and when to retry](https://broadrouter.com/guides/errors-and-retries/)

Try it on your real workload. New accounts start with $5 of credit for Gemma and transcription. DeepSeek, GLM and Qwen3.8 need a top-up.

[Get started](https://inference.broadnet.ai/account/?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-arabic-speech-to-text-api-buyers-guide&utm_content=page-cta) ·

**Sources**

-   [BroadNet Transcribe Arabic model page: head-to-head WER against ElevenLabs Scribe v2 and Whisper large-v3 (fetched 2026-11-15)](https://www.broadnet.ai/inference/models/cohere-transcribe-arabic)
-   [BroadNet model catalogue, speech to text: WER and leaderboard rank as reported by BroadNet (fetched 2026-11-14)](https://www.broadnet.ai/inference/models)
-   [Open Universal Arabic ASR Leaderboard, Hugging Face Space (fetched 2026-11-15)](https://huggingface.co/spaces/elmresearchcenter/open_universal_arabic_asr_leaderboard)
-   [BroadNet Inference API reference: transcription formats, 26,214,400-byte limit, 413, language field (fetched 2026-11-14)](https://inference.broadnet.ai/docs?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-arabic-speech-to-text-api-buyers-guide&utm_content=sources)
-   [BroadNet Inference machine-readable prices, pricing.json (fetched 2026-11-14)](https://inference.broadnet.ai/pricing.json?utm_source=broadrouter&utm_medium=referral&utm_campaign=blog-arabic-speech-to-text-api-buyers-guide&utm_content=sources)
-   [Soniox pricing: async about 0.10 USD/hour, real-time about 0.12 USD/hour, billed per token (checked 2026-11-14)](https://soniox.com/pricing)
-   [Soniox docs, supported languages: Arabic (ar) listed among 60 (checked 2026-11-14)](https://docs.soniox.com/stt/concepts/supported-languages)
-   [jiwer, Python package for ASR error rates (fetched 2026-11-14)](https://github.com/jitsi/jiwer)
