Romanian transcription: how the output reads

By the Potok teamUpdated ~2 min read

To see Romanian transcription output, send a short clip to Potok on Telegram. The text returns in Romanian, with a speaker label and a time if you choose them. Published tests of large-v3 report a word error rate of 8.2, and 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

  • The 8.2 is a published word error rate of others, from large-v3 tests, not our measurement.
  • Romanian has five extra letters: a with breve, a with circumflex, i with circumflex, s and t with a comma below.
  • Two sources of published figures agree that Romanian is in the recommended group.
  • The bot returns Romanian only and does not add a version in another language.

What a transcript looks like

Sample sentence in Romanian
[00:00] Speaker 1: Să începem ședința. Mai întâi vom analiza rezultatele săptămânii trecute, apoi vom planifica următorii pași.

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
Romanian (Română)Recommendedword error rate 8.2% (large-v3, FLEURS)
Canary-1B-v2 paper, Whale 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. Try a clip. Message @potok_now_bot with a voice note or short video.
  2. Look at the layout. Notice how each speaker turn is placed, and whether times help you.
  3. Send the real recording. A file up to 2 GB goes straight in; for the largest, tap Upload in the mini app.
  4. Verify the special letters. Skim for the five extra letters and confirm they are used consistently.
  5. Save it. Choose TXT, MD, DOCX, PDF, SRT or VTT.
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
Mixing two forms of the same letterSome fonts or older documents use a cedilla in place of a comma below, so search gets inconsistent.Normalise to one form in your editor.
Skipping a sample clipYou discover layout surprises only after a two-hour upload.Test a minute first.
Reading the 8.2 as a promiseYour audio may be noisier than the test sets.Treat it as a published reference point.
Expecting names with the right caseForeign names may be spelled by sound.Check names against a source.

More

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

Frequently asked questions

How does a Romanian transcript look?

It is Romanian text split by speaker, with optional times at the start of each paragraph.

Is the diacritic spelling kept?

The text uses the Romanian letters, but check that your font and editor display them properly.

What is the published error rate?

A word error rate of 8.2 for large-v3 from published tests by others.

Can I export SRT?

Yes, along with VTT, TXT, MD, DOCX and PDF.

Does the bot translate Romanian?

No. It writes down what was said, in Romanian.

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. Canary-1B-v2 paper, Table 11 (Whisper large-v3 on FLEURS) · checked
  4. Whale paper, Table 5 (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