Convert Arabic audio to text

By the Potok teamUpdated ~3 min read

To convert Arabic audio to text, send the file to the Potok bot on Telegram, with one caution. Published tests of others on the large-v2 model, not our measurements, report a word error rate of 16 for Arabic, and dialect speech does worse. Files up to 2 GB are accepted.

Transcribe a recording

Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.

Open @potok_now_bot in Telegram

Files up to 2 GB · export to TXT, MD, DOCX, PDF, SRT and VTT · speaker labels and timestamps · about 30 languages

Key facts

  • A turbo average of 10.57 from a third party shows that published Arabic figures disagree with each other.
  • Colloquial dialects sound far from Modern Standard Arabic, and published error rates are higher on them.
  • Text is written right to left, so the exported file needs an editor that keeps that direction.
  • Newsreaders and lecturers speak close to the standard form, so their recordings suit the number best.

What a transcript looks like

Sample sentence in Arabic
[00:00] Speaker 1: لنبدأ الاجتماع. أولاً سنراجع نتائج الأسبوع الماضي، ثم نخطط للخطوات التالية.

Meaning in English: Let's start the meeting. First, we'll review last week's results, then plan the next steps.

Written by us to show the script and the layout of a transcript. It is not the output of the recognition engine.

Published quality

Quality for this language
LanguageVerdictPublished resultNotes
Arabic (العربية)Use with careword error rate 16% (large-v2, FLEURS)
Whisper paper, TheWhisper benchmark
Published tests show a higher error rate for this language. Check a short sample first and plan to proofread.
The figure is from the earlier large-v2 model and a third-party turbo average was 10.57. Dialects differ a lot from Modern Standard Arabic, and error rates are higher on dialect speech.

These numbers come from published tests by other people, not from tests by Potok. Your own audio, with its noise, accents and microphone, will differ.

How to do it

  1. Decide which Arabic you recorded. Prepared speech such as a bulletin or a lecture sits near the standard form. A family call or a market chat is dialect. This tells you how closely to proofread later.
  2. Open the bot. Open @potok_now_bot in Telegram and pick the chat.
  3. Hand over the recording. Attach audio, a video, a voice note or paste a YouTube link. Anything up to 2 GB works; for a heavy file press Upload in the bot's mini app.
  4. Test the opening minute. Read the first minute of text while the audio plays. Wrong names or numbers mean you should read the whole transcript before sharing it.
  5. Save the result. Pick TXT or DOCX to read, SRT or VTT for subtitles. Keep timestamps on if you will return to the audio.
What Potok can do
FeatureDetails
Where it worksTelegram bot @potok_now_bot; large files go through the bot's mini app
File sizeUp to 2 GB
ExportTXT, MD, DOCX, PDF, SRT, VTT
Speakers and timeSpeaker labels and timestamps can be switched on
LanguagesAbout 30 languages with good published results; a few more with warnings
TranslationThe transcript is not translated into another language
DictationA separate Potok desktop app lets you dictate text into any program; it recognises Russian speech

Common mistakes

Common mistakes and how to fix them
MistakeWhat happensFix
Applying one score to every dialectColloquial speech yields clearly more errors than formal speech.Run a short clip from the same speaker first.
Opening the file in a left-to-right editorDigits and punctuation jump to the wrong end of a line.Use a word processor with right-to-left support.
Presenting the 16 as our own resultReaders assume it was measured on their recording.State that it comes from published tests of others.
Expecting short vowel marksEveryday Arabic often leaves them out, so a missing mark is not always an error.Add marks by hand if your reader needs them.

More

See the full list of languages, the tools and the glossary.

Frequently asked questions

Why does the Arabic page carry a caution?

Published error rates for Arabic are higher than for the recommended languages, and they rise on dialect speech, so the output deserves a careful read.

Does it handle Egyptian, Gulf or Levantine speech?

The published tests do not split results by dialect. They only note that dialects differ a lot from Modern Standard Arabic, so try a short sample first.

Can I send an Arabic video instead of audio?

Yes. The bot reads video files and YouTube links as well as audio, within the 2 GB limit.

Which formats can I export?

TXT, MD, DOCX, PDF, SRT and VTT are available, and speaker labels or timestamps can be switched on before you send.

Does the bot turn the Arabic into another language?

No. The text stays in the language of the recording.

Sources

Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.

  1. Word error rate (Wikipedia), Wikipedia · checked
  2. FLEURS speech dataset (Google, Hugging Face), Hugging Face · checked
  3. Whisper paper, Table 13 (large-v2 on FLEURS) · checked
  4. TheWhisper benchmark (turbo, average over FLEURS, MLS, Common Voice) · checked

Try it on your own recording

Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.

Open @potok_now_bot in Telegram