To check a free German transcription tool, send a clip with compound nouns and formal names, then compare each with the audio. Potok cites published large-v3 word error rates of 4.30 to 5.8 for German, from other people's tests. Read the result against the audio before you trust a long file.
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 German range from 4.30 to 5.8 in word error rate.
- The figures come from published tests by others, not our measurements.
- German writes compound nouns as one word and capitalises every noun, so a text check reveals care.
- The transcript stays in German and is not translated.
What a transcript looks like
[00:00] Speaker 1: Lassen Sie uns die Besprechung beginnen. Zuerst gehen wir die Ergebnisse der letzten Woche durch, danach planen wir die nächsten Schritte.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 |
|---|---|---|---|
| German (Deutsch) | Recommended | word error rate 4.3% (large-v3, FLEURS) Canary-1B-v2 paper, Whale paper, TheWhisper benchmark | Published tests show a low error rate for this language. Published large-v3 results range from 4.30 to 5.8. |
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
- Collect compounds. Use a clip with long compound nouns, such as those in contracts or technical talks.
- Open the bot. Open @potok_now_bot in Telegram and send the clip as a voice message or an audio file.
- Check capitals. Every noun starts with a capital letter. See whether the tool follows that.
- Check formal names. Compare titles, surnames and company names with the audio.
- Move on to the real file. Up to 2 GB is accepted, with Upload in the mini app for large files. Export TXT, DOCX or PDF, and keep the short clip's result so you can compare it with the long file.
| 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 |
|---|---|---|
| Ignoring split compounds | A noun cut into two parts changes the reading. | Mark each split as an error. |
| Overlooking lowercase nouns | It signals sloppy punctuation and capitalisation. | Scan for lowercase nouns. |
| Quoting only 4.30 | Other tests give up to 5.8. | Quote the range and say which model. |
| Testing with dialect speech and expecting standard text | Regional speech is a different task. | Use standard speech first. |
More
See the full list of languages, the tools and the glossary.
Frequently asked questions
How does German score?
Published large-v3 results range from 4.30 to 5.8 in word error rate.
Why compound nouns?
They are long, and a tool that splits them gives text that looks wrong.
Can I pick the export format?
Yes: TXT, MD, DOCX, PDF, SRT or VTT.
Do you handle YouTube links?
Yes, along with files and voice messages, within the 2 GB limit.
Does the text change language?
No. The transcript stays in the language of the recording.
How long should a German test clip be?
About one minute is enough. It should contain several long nouns, a few names and at least one number, so the check covers the hard cases.
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
- TheWhisper benchmark (turbo, average over FLEURS, MLS, Common Voice) · 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