Kalemio
Menu
العربية

Arabic speech-to-text API

Arabic dialects.
Every word, timed.

Transcribe Arabic without automatically rewriting the dialect into formal Arabic. Get word-level timestamps, speaker labels, JSON, and SRT for captions and searchable transcripts.

Get an API keyRead the API guide

Starting at 200 minutes for USD 1.26 /month. Word timestamps, speaker labels, JSON, and SRT included.

Fewer transcription errors.

Dialect-tolerant word error rateLower is better

  1. Kalemio4.6%
  2. Meta Muse Voice 1.06.3%
  3. Hamsa General V27.4%
  4. ElevenLabs Scribe v27.4%
  5. Groq Whisper large-v317.0%
  6. Groq Whisper turbo25.1%

Kalemio benchmark · September 2026. Same audio and scoring for every model, with spelling and equivalent word forms normalized. Results vary by recording.

The words matter. So does the dialect.

Interviews, podcasts and everyday conversations don’t sound like a textbook. Kalemio is built for Modern Standard Arabic and spoken dialects, keeping the words as they were said.

A timestamp for every word.

Each word in the JSON response includes a start time, end time, and speaker label. Highlight speech during playback, jump from search results to the right moment, or build your own caption layout.

Use JSON for word-level timing and custom interfaces. Download SRT for subtitle cues you can import into an editor or player.

Explore word-level timestamps
A little Levantine Arabic

أهلاً، كيف فيني ساعدك؟

“Hello, how can I help you?”

  • أهلاً،0.24–0.68
  • كيف0.72–0.94
  • فيني0.98–1.22
  • ساعدك؟1.26–1.78

Illustrative transcript. Times are in seconds.

See the JSON
{ "text": "أهلاً، كيف فيني ساعدك؟",
  "words": [
    {"text":"أهلاً،","start":0.24,"end":0.68,"speaker_id":"speaker_0"},
    {"text":"كيف","start":0.72,"end":0.94,"speaker_id":"speaker_0"},
    {"text":"فيني","start":0.98,"end":1.22,"speaker_id":"speaker_0"},
    {"text":"ساعدك؟","start":1.26,"end":1.78,"speaker_id":"speaker_0"}
  ]
}

Build captions, search, and interactive transcripts.

Arabic subtitles

Download SRT for your video editor, or use word timestamps to make captions your own.

Podcasts and interviews

Keep the original dialect and follow the conversation with speaker labels.

Searchable recordings

Link each word to its moment so listeners can find a passage and go straight to it.

Just need a subtitle file?

Arabic Subtitles is a separate no-code product for creating SRT files. These guides cover subtitle workflows; the Kalemio API remains the product for software integrations.

Generate an Arabic SRT without writing code

Arabic subtitle guides

Plans renew monthly until cancelled. Monthly minutes reset at renewal and don’t roll over. Upgrade or cancel from your account.

Every plan supports files up to 10 minutes and 20 MiB, with 2 active jobs per account.

Add 100 extra minutes.

100 extra minutes for USD 1.08. A one-time purchase for subscribers. These minutes never expire, even if you cancel your plan.

Send the audio. Or the video.

MP3, M4A, PCM WAV, MP4 and MOV. Up to 20 MiB and 10 minutes per file, with two active jobs per account.

File formats and limits

About the Arabic transcription API

Your recordings stay yours.

We don’t use them to train models. Working audio is deleted after processing; results are yours to download for 24 hours after submission.

How we handle your data

Before you start

A few things to know about transcription, timing and the API.

Does Kalemio support Arabic dialects?

Yes. Kalemio is built for Modern Standard Arabic and spoken dialects, keeping dialect words instead of rewriting them into formal Arabic. See the benchmark on this page for the current evaluation scope.

What are word-level timestamps?

Each word in the JSON transcript has a start and end time in seconds, plus a speaker ID. Use these times to highlight words during playback or jump to the exact passage in a recording.

Can I create Arabic subtitles?

Yes. Download SRT subtitles to add to your video, or use the words and timestamps in JSON to build and style your own captions.

Which files can I upload?

MP3, M4A, PCM WAV, MP4 and MOV, up to 20 MiB and 10 minutes per file. You can send video directly; we extract the audio for you.

How do I get started?

Create an account, choose a monthly plan, and create an API key. Upload your file, check its progress, then download the completed transcript. The API guide includes examples for each step.

Are my recordings used to train models?

No. We delete working audio after processing. Results are available to download for 24 hours after submission, so save a copy of anything you need.

Have Arabic audio to work with?

Create an account, choose a plan, and get your API key. The guide walks you through your first request. Just need a subtitle file? Arabic Subtitles is a separate no-code product for generating an Arabic SRT.

Get an API key