Skip to content
English
Start free
Open menu

Speech to textPaid plans

Arabic meeting transcription with speakers and timestamps

Upload your meeting recording once the meeting is over and get the full text, attributed to each speaker, with a time on every line. For teams whose meetings run in Arabic and English together and who need a written record to go back to when writing minutes and following up.

Get StarterGet BasicSee the output
  • Speakers detected automatically
  • Word-level timestamps
  • Word, SRT, VTT, TXT, JSON and CSV

Included in every paid plan, from $1/month (Basic, Starter, Creator, Professional, Business and Enterprise)from $4.99/month (Starter, Creator, Professional, Business and Enterprise). Not available on the Free plan.

See the workflow

A real transcript and its translation

Speaker 1Speaker 2Detected: English
0:00 / 0:36

Transcript

Real output, unedited · voices by Vocla
Two-speaker clips voiced with Vocla library voices, uploaded and transcribed by Vocla with language and speakers on auto. Lines, speakers, word timings and the translation are exactly as returned.

What you get

What a meeting transcript looks like

An excerpt from a meeting transcript after the speakers were named, a names list to paste before transcribing, and a minutes template you fill in from the text. All names are invented.

  • Sample 1

    Excerpt from a weekly meeting

    Saudi Arabic

    [00:12:05] نورة: طيب وش صار على الـ landing page؟ [00:12:09] فيصل: التصميم جاهز، بس المحتوى الإنجليزي ما وصلنا للحين. [00:12:16] نورة: أبغاه يوم الأربعاء، لأن الحملة تبدأ الأحد. [00:12:21] فيصل: تمام، أرسله لك الأربعاء الظهر.

    Noura: "So what happened with the landing page?" Faisal: "The design is ready, but the English copy hasn't reached us yet." Noura: "I want it on Wednesday, because the campaign starts Sunday." Faisal: "Fine, I'll send it to you Wednesday at noon." Each line shows the time, the speaker's name and the words as spoken.

  • Sample 2

    Names and terms to paste before transcribing

    Modern Standard Arabic

    نورة، فيصل، ريم، مشروع واحة، عميل قطاع التجزئة، الربع الثالث، خطة الإطلاق، KPI، landing page، sprint، ميزانية الحملة

    Noura, Faisal, Reem, Project Waha, the retail client, third quarter, launch plan, KPI, landing page, sprint, campaign budget. Spell each name the way you want it in the minutes and keep English terms in Latin letters.

  • Sample 3

    Minutes template to fill in from the text

    Modern Standard Arabic

    محضر اجتماع فريق التسويق. الحاضرون: نورة، فيصل، ريم. القرار الأول: اعتماد تصميم الصفحة كما عُرض (الدقيقة 12). المهمة: يسلّم فيصل المحتوى الإنجليزي يوم الأربعاء. الاجتماع القادم: الأحد الساعة العاشرة صباحًا.

    Marketing team meeting minutes. Present: Noura, Faisal, Reem. Decision one: the page design is approved as shown (minute 12). Task: Faisal delivers the English copy on Wednesday. Next meeting: Sunday at ten in the morning. Put the minute next to each decision so anyone can go back and listen.

The problem

The meeting is over and nobody remembers who promised what

An hour of discussion between six people, half of them speaking dialect and dropping English words like deadline and budget into every sentence.

  1. The person asked to take notes got pulled into the discussion and left with half a page that has no figures and no record of who committed to deliver.

  2. The recording exists, but going back to it means listening to the whole hour for one sentence said at minute forty. When two people disagree a week later about what was decided, nobody has time to play the file again.

  3. A team with colleagues who do not read Arabic has one more problem: a discussion held in Arabic needs someone to summarise it for them out loud, so they receive a thinner version of what was said.

Why Vocla

What Vocla does for meeting recordings

  • Every line attributed to its speaker

    Speakers are detected automatically and shown as Speaker 1 and Speaker 2. Type the real name once and it changes across the transcript, and if a line lands on the wrong person you move it in the editor.

  • Timestamps that take you back to the moment

    Every line carries its time in the recording. Click the sentence in dispute and hear it from that point, with no scrubbing back and forth through an hour-long file.

  • A names-and-terms list before you transcribe

    Colleagues, clients, project names and internal abbreviations are where any speech recogniser slips most. Add them to a list before you start and they are written the way your company writes them.

  • Dialect and English in the same recording

    You choose Arabic and the speakers' dialect when you upload. Quiet meetings with a close microphone give a more accurate text; dialect over room noise needs more review, which the editor makes quick.

  • Side-by-side translation and Word export

    Translate the transcript from Arabic to English or the other way with every line keeping its time, and export a Word file with right-to-left Arabic paragraphs containing the original, the translation or both.

How it works

From a meeting recording to a written record

  1. Step 1

    Save the recording and upload it afterwards

    Download the audio or video file from your meeting software or the recorder you used and upload it to Vocla. Files up to 500 MB are accepted. Transcription runs on uploaded files, not during the meeting.

  2. Step 2

    Paste the names list and choose the dialect

    Add the attendees, the project, the client and recurring terms, and set the language and dialect yourself instead of leaving it on auto when the conversation moves between two languages.

  3. Step 3

    Name the speakers and correct what matters

    Replace speaker numbers with names, then go over figures, dates and names, which are what minutes depend on. Find and replace fixes a recurring name across the whole file at once.

  4. Step 4

    Write the minutes from the transcript

    Export the Word file and open it next to your minutes template. You write the decisions and tasks from the text: the tool transcribes what was said and does not produce summaries or action items automatically.

  5. Step 5

    Share the right version with each colleague

    Send the bilingual file to those who read one language, and keep the transcript in the workspace, where viewers can open, search and export it.

Cost

What does transcribing a meeting cost?

Speech to text is available on paid plans and is charged by the minute: 150 credits per started minute of the recording, plus 60 credits per minute if you ask for a translation. The cost is shown before you start, based on the file's length, and the credits are refunded if the transcript fails.

The longest file Starter accepts is 30 minutes, enough for short meetings. Meetings of an hour or more suit Professional, with its 120 minute limit per file and three team seats, and Business takes files up to 300 minutes.

Compare plans

The numbers

Quick maths

Worked example: a 60-minute recording is 9,000 credits, or 12,600 with translation.

  • Not available on the Free plan; included in every paid plan.
  • Transcription costs 150 credits per started minute of audio, shown before you start and refunded if the transcript fails.
  • Translating the transcript adds 60 credits per minute.
  • Longest file: 30 minutes on Starter, 120 on Professional and 300 on Business.

Figures use the current plans and default credit rates. The exact cost is always shown in the app before you confirm.

Questions people ask

How do I turn a meeting recording into text?

Save the meeting recording as an audio or video file, upload it to speech to text in Vocla, set the language and dialect and add the attendees' names. Usually within a few minutes the transcript appears with speakers and timestamps; review it and export it to Word.

Does it write the minutes and a summary automatically?

No. Vocla writes out what was said as a full transcript with speakers and times, and does not generate a summary or a task list. You write the minutes from the text, which is much faster than writing from memory or listening again, because you search for a word and jump straight to it.

Does it join the meeting and record it live?

No. There is no assistant that joins the meeting and no live transcription while people talk. Record the meeting the way your team usually does, after telling attendees it is being recorded, then upload the file when it ends.

Does it understand Arabic mixed with English?

Yes, English words inside Arabic speech are written down, and results improve if you add recurring terms to the names list before transcribing. Accuracy on dialects is still lower than on Modern Standard Arabic when the recording is noisy, so set aside time to check figures and names.

How many speakers can it tell apart in a meeting?

Speakers are detected automatically and numbered in the order they appear. Detection works well when people take turns; two similar or overlapping voices can be merged into one speaker. When that happens, split the line and move it to the right person in the editor.

Can I translate the meeting transcript into English?

Yes. Lines are translated one by one, keep their times and appear next to the original, and you can export the original, the translation or a bilingual file. Translation adds 60 credits per minute to the cost of the transcript.

Go deeper

The broader use case

Related solutions

Other pages for a neighbouring job.

All solutions

Start in minutes

Give your content a voice today

Start free with ⁠500͏⁠ credits every month. No card required.

Create your free accountSee pricing
  • ⁠500͏⁠ free credits every month
  • No card required
  • Arabic and English interface