Send a Romanian recording to the Potok Telegram bot to convert Romanian audio to text. Published tests of others on the large-v3 model, not our measurements, report a word error rate of 8.2 for Romanian. Short results arrive in the chat, longer ones as a file, up to 2 GB per upload.
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
- Two published sources agree on the large-v3 result, which makes 8.2 a little steadier than a single figure.
- Romanian has five marked letters, and two of them carry a mark below the letter that some old fonts draw badly.
- The definite article is attached to the end of the noun, so a wrong ending changes the meaning.
- Informal speech often drops syllables, and the text may keep them dropped.
- Some speakers drop the final vowel sounds, so a clipped ending is a prime spot to proofread.
What a transcript looks like
[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
| Language | Verdict | Published result | Notes |
|---|---|---|---|
| Romanian (Română) | Recommended | word 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
- Prepare the audio. Trim long silences at the start if you can. They add nothing to the transcript.
- Open the bot. Open @potok_now_bot and attach the file.
- Send a large recording. Anything up to 2 GB goes through, and Upload in the mini app is there for files the chat rejects.
- Inspect the marked letters. Check that the two letters with a mark below appear in the form your readers expect.
- Read noun endings. Compare a short stretch with the audio, looking for article endings.
- Export. Choose DOCX or TXT for editing, SRT or VTT for subtitles.
| Feature | Details |
|---|---|
| Where it works | Telegram bot @potok_now_bot; large files go through the bot's mini app |
| File size | Up to 2 GB |
| Export | TXT, MD, DOCX, PDF, SRT, VTT |
| Speakers and time | Speaker labels and timestamps can be switched on |
| Languages | About 30 languages with good published results; a few more with warnings |
| Translation | The transcript is not translated into another language |
| Dictation | A separate Potok desktop app lets you dictate text into any program; it recognises Russian speech |
Common mistakes
| Mistake | What happens | Fix |
|---|---|---|
| Mixing old and new marked letters | Cedilla and comma forms look similar but are different characters. | Run a search for the form you want. |
| Not checking endings | The article sits on the noun. | Compare with the audio. |
| Reading 8.2 as exact | It is an average from others' tests. | Proofread names. |
| Forgetting speaker labels in a debate | Voices run together. | Switch labels on. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
What does 8.2 mean?
It is a published word error rate on large-v3: about eight wrong words per hundred in those tests.
Do I need a special keyboard to edit?
No. Any editor with Unicode support will show the Romanian letters.
Is the limit 2 GB?
Yes, per file, with the Upload button for large ones.
Are timestamps possible?
Yes, switch them on before sending and export SRT or VTT.
Is the Romanian translated?
No. The transcript stays in Romanian, the same language as the recording you send, and it is never converted into another one.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- Word error rate (Wikipedia), Wikipedia · checked
- FLEURS speech dataset (Google, Hugging Face), Hugging Face · checked
- Canary-1B-v2 paper, Table 11 (Whisper large-v3 on FLEURS) · checked
- 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