To check a free Czech transcription tool, send a clip with long words and numbers and compare each line with the audio. Potok cites published tests of large-v3 with word error rates from 11.0 to 11.33 for Czech, and those are other people's numbers.
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
- Published large-v3 results for Czech range from 11.0 to 11.33 in word error rate.
- These figures come from other people's tests, not our measurements.
- Accent marks and long words make Czech a useful language for spotting careless output.
- The transcript stays in Czech and is not translated.
What a transcript looks like
[00:00] Speaker 1: Začněme schůzku. Nejdříve projdeme výsledky z minulého týdne a potom naplánujeme další kroky.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 |
|---|---|---|---|
| Czech (Čeština) | Recommended | word error rate 11% (large-v3, FLEURS) Canary-1B-v2 paper, Whale paper | Published tests show a low error rate for this language. Published large-v3 results range from 11.0 to 11.33. |
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
- Build a tricky clip. Read a paragraph with long words, a phone number and a street address.
- Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message or an audio file.
- Inspect the accents. Czech uses marks above many letters. A tool that drops them gives poor text even when sounds are right.
- Verify numbers. Compare every figure with the audio, since digits are easy to mishear.
- Use the long-file route. Files up to 2 GB work, and Upload in the mini app covers large ones. Export TXT, DOCX or PDF.
| 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 |
|---|---|---|
| Skipping the accents check | Words with missing marks look plausible but are wrong. | Scan for unmarked words in the output. |
| Reading only the first lines | Quality can change over a long recording. | Spot-check the middle and end too. |
| Reporting only the best published number | The range runs from 11.0 to 11.33. | Quote the whole range. |
| Using a clip with overlapping voices | Interruptions blur the comparison. | Use a single clear speaker first. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
What do published tests say about Czech?
Large-v3 results range from 11.0 to 11.33 in word error rate, which counts as a good result.
Is the figure from your own tests?
No. It comes from published tests by others.
Can I add speaker labels?
Yes. Speaker labels can be switched on, and they appear as numbers such as Speaker 1.
What export formats are there?
You can save the transcript as TXT, MD, DOCX, PDF, SRT or VTT, so a plain text file and subtitles are both covered.
Is there a version in English?
No. The transcript stays in the language of the recording.
Should I test with a phone recording?
Yes, if that is how your real files are made. A phone recording shows how the tool copes with room noise and uneven volume.
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
- 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