Free Croatian transcription: test a tool on a short clip

By the Potok teamUpdated ~2 min read

To test a free Croatian transcription tool, send a short clip and check how it handles the special letters. The published word error rate for Croatian with large-v3 is 11.07 in a test by others, and Potok treats it as a guide, not a promise for your file.

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

  • The 11.07 is a word error rate for large-v3 in a published test, not our measurement.
  • Croatian uses Latin letters with extra marks, and a tool that drops them changes the words.
  • Case endings change word forms, so one wrong ending still counts as a wrong word.
  • The transcript stays in Croatian and is not translated.

What a transcript looks like

Sample sentence in Croatian
[00:00] Speaker 1: Počnimo sastanak. Prvo ćemo pregledati rezultate prošlog tjedna, a zatim isplanirati sljedeće korake.

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
Croatian (Hrvatski)Recommendedword error rate 11.07% (large-v3, FLEURS)
Canary-1B-v2 paper
Published tests show a low error rate for this language.

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. Choose a clip with special letters. Pick a minute of speech with words that carry the extra marks, such as common place names.
  2. Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message or an audio file.
  3. Read for diacritics. See whether the marked letters survive. A missing mark often turns one word into another.
  4. Check word endings. Compare the endings with what the speaker says; they carry the grammar and are easy to get wrong.
  5. Send the real file. Up to 2 GB is accepted, with the Upload button in the mini app. Export TXT or DOCX for editing.
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
Accepting text with missing marksA word may still look plausible while meaning something else.Search the output for words that should carry marks.
Overlooking wrong endingsThe sentence reads smoothly but the grammar is off.Read slowly alongside the audio.
Testing on a mumbled interviewPoor sound blurs the comparison between tools.Start with a clear recording.
Taking 11.07 as a promiseIt was measured on other audio.Treat it as a guide and rely on your own clip.

More

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

Frequently asked questions

How well is Croatian covered?

Published tests give a word error rate of 11.07 for large-v3, which is in the recommended range.

Why do the extra letters matter?

A dropped mark can change a word, so they are a quick probe for output quality.

Can I send a YouTube link?

Yes. The bot accepts YouTube links as well as files, within the 2 GB limit.

Can I get subtitles?

Yes. Export SRT or VTT from the finished transcript.

Does the text come back in another language?

No. The transcript 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. FLEURS speech dataset (Google, Hugging Face), Hugging Face · checked
  2. Word error rate (Wikipedia), Wikipedia · checked
  3. Canary-1B-v2 paper, Table 11 (Whisper large-v3 on FLEURS) · 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