Free Vietnamese transcription depends on tone marks, so test Potok with a short clip. Published tests of others on the earlier large-v2 model give a word error rate of 10.3 for Vietnamese, and another test set gave 18.9 on large-v3. Read a sample against the audio and look at each mark.
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 scores vary between test sets, which means 10.3 is a guide and not a prediction.
- Both figures come from other people's benchmarks, not our measurements.
- Vietnamese is written with tone and vowel marks, and a wrong mark makes a different word.
- Each syllable is written as its own word, so a single wrong syllable is a wrong word.
What a transcript looks like
[00:00] Speaker 1: Chúng ta bắt đầu cuộc họp nhé. Trước tiên, chúng ta sẽ xem lại kết quả tuần trước, sau đó lên kế hoạch cho các bước tiếp theo.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 |
|---|---|---|---|
| Vietnamese (Tiếng Việt) | Recommended | word error rate 10.3% (large-v2, FLEURS) Whisper paper | Published tests show a low error rate for this language. The figure is from the earlier large-v2 model; another test set gave 18.9 for large-v3, so results vary with the data. |
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
- Pick speech with tones. A clip of ordinary conversation tests the marks better than a list of words.
- Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message, an audio file or a video.
- Check tone marks. Compare marked vowels with the audio and tick the matches.
- Test a second voice. Try a speaker from a different region, since accents change how tones sound.
- Export the result. Save TXT, DOCX, PDF, SRT or VTT. Files up to 2 GB are accepted.
| 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 |
|---|---|---|
| Accepting a missing mark | It makes a different word. | Proofread marks. |
| Trusting a single figure | 10.3 and 18.9 differ widely. | Test your own audio. |
| Testing one region | Northern and southern speech differ. | Add a second speaker. |
| Reading in a font without marks | Marks display badly. | Use a full-coverage font. |
| Judging by a single syllable | Each syllable is its own written word, so one slip counts as a whole wrong word. | Count syllables, not sentences, when you tally. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
Why do two figures differ so much?
They come from different models and test sets, so 10.3 and 18.9 are both published but not alike.
Which model do they refer to?
10.3 is for large-v2 and 18.9 is for large-v3.
Is it your own measurement?
No. It comes from published tests of others.
Should I test more than one clip?
Yes. A northern and a southern voice show how much the tones matter for the tool you are judging.
Can I get English text?
No. The transcript stays in the language of the recording.
What is the file size limit?
Files up to 2 GB are accepted, and a long result arrives as a file you can save.
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
- Whisper speech recognition system (Wikipedia), Wikipedia · checked
- Whisper paper, Table 13 (large-v2 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