Sinhala audio

Transcribe Sinhala audio to text

Transcribe Sinhala audio to text: our engine writes down 99+ languages, AssemblyAI lists Sinhala among 99 languages. Where each part of a recording should go.

3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps
In short

Sinhala is not among our 99+ languages: the engine handles 99+ languages. AssemblyAI publishes Sinhala in a list of 99 languages (their language page, read 22 October 2026). Give the Sinhala recording to them; the English part of that session comes here, up to 30 minutes.

The case

A session with two voices and 99+ languages

A workshop recording is usually more than one language: the teacher speaks Sinhala, the questions from the visiting colleague come in English, and the answers go back into Sinhala. The recording language is detected automatically.

The numbers

What one session can become

Sinhala speech

on our list — our engine has 99+ languages

A service that lists Sinhala

AssemblyAI lists 99 languages, Sinhala among them

Speakers

named separately, on request

Export formats

TXT, Markdown, SRT, VTT, CSV, and DOCX on Pro

How it works

How to handle a session in 99+ languages

  1. 1

    Find the English in the session

    The questions from abroad and the summary at the end: that part is English, and it is the part our engine takes.

  2. 2

    Send the Sinhala half elsewhere

    AssemblyAI lists Sinhala among the 99 languages on its language page, read on 22 October 2026. The link is at the bottom of this page.

  3. 3

    Upload the English half here

    Drop the file on the widget above. It may be up to 30 minutes, and the speakers can be named on request.

  4. 4

    Export for the people who will read it

    TXT for the archive, Markdown for the notes, DOCX on Pro for the editor. Every format keeps the timecodes of the lines.

Tips

  • Cut the session by the language, not by the hour.

  • Ask for the split by speaker on a room recording.

  • Keep the slides; the transcript is only the words.

  • Try the free plan on the English half first.

Questions this page answers

Will Sinhala speech be recognised by mistake?

Yes, and that is the trap.Tthe recording language is detected automatically.

Can the speakers be told apart?

Yes, switch on the split by speaker before the upload. A room with a teacher and a dozen participants becomes lines with names instead of one long column.

What do the exports give me?

TXT, Markdown, SRT, VTT and CSV on every plan, and DOCX on Pro. The subtitle files carry the same start and end times as the transcript.

How long can the recording be?

up to 30 minutes for one file, on every plan. A longer session has to be cut into parts before the upload; the service refuses the file with a clear message.

Where does the audio go?

Athe recording language is detected automatically. Nothing is used to train a model for anybody else.

Related pages

Languages we can transcribe

Which languages our engine takes, with the code of each and whether the language is detected automatically.

Transcribe English audio to text

The page about the language we do support.

Transcribe Burmese audio to text

The same question for another language.

Transcribe a podcast episode to text

Where a session recording ends up.

Sources

  1. A

Reading: 6 minutesUpdated Transcribsi editorial team

Drop a recording — the text arrives in a minute

Speakers by name, a time on every line, export to TXT, SRT, VTT, CSV and Markdown.

3 files a day up to 30 minutes, free99+ languagesSpeaker separation and timestamps