Vietnamese speech to text in Potok is a recommended language, with a published word error rate of 10.3 from the earlier large-v2 model. Another test set gave 18.9 for large-v3, so results vary with the data. These are other people's tests, not our measurements.
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 gap between 10.3 and 18.9 is large, so expect your recording to fall anywhere in a wide band.
- Six tones are marked with diacritics, and a wrong mark turns one word into another.
- Each syllable is written as a separate word with spaces, which makes word error rate count many short units.
- Northern and southern speech differ in sound, and the published figures do not split them.
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
- Choose a test clip. Pick one minute with both ordinary talk and a few names.
- Send to the bot. Open @potok_now_bot and share the file, a voice message or a YouTube link.
- Mind the limit. Files up to 2 GB are accepted; Upload in the mini app suits the largest.
- Check tone marks. Compare a sample of words with the audio, focusing on those where a mark changes the meaning.
- Check names. Personal and place names often carry less common mark combinations.
- Export. TXT, MD, DOCX, PDF, SRT or VTT.
| 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 |
|---|---|---|
| Quoting only 10.3 | Another test gave 18.9. | Show both and say they use different data. |
| Skipping the tone check | A wrong mark can make another word. | Spot-check every page. |
| Using a recording with a heavy regional accent without a test | Published figures do not split regions. | Run a short test. |
| Pasting into an editor that strips diacritics | Meaning is lost. | Use UTF-8 files or DOCX. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
What are the published scores?
10.3 for large-v2 in one test, and 18.9 for large-v3 in another test set.
Why do the numbers differ so much?
They come from different data and different models, so they cannot be compared directly.
Whose tests are they?
They are published by others, not measured by us.
Does the bot handle northern and southern speech?
Published figures do not split accents, so check a short clip from your speaker.
Does the text stay in Vietnamese?
Yes. It is not changed into another language.
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
- 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