Convert German audio to text

By the Potok teamUpdated ~2 min read

Potok converts German audio to text when you send the file to its Telegram bot. Published tests of others on large-v3, not our measurements, show a word error rate of 4.3, with results from 4.30 to 5.8. Files up to 2 GB are accepted.

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 range of results shows that different tests disagree by more than one point, so treat the headline as a guide.
  • German compounds can come out as one word or split into two, which a reader may take for a typo.
  • Nouns carry capital letters, so check them in the output.
  • Swiss and Austrian speech can differ from standard German in sound and vocabulary.

What a transcript looks like

Sample sentence in German
[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

Quality for this language
LanguageVerdictPublished resultNotes
German (Deutsch)Recommendedword 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

  1. Open the bot. Open @potok_now_bot in Telegram.
  2. Send the file. Audio, video, a voice note or a YouTube link, up to 2 GB. Use the Upload button for big files.
  3. Check compounds. A long compound may be split into two words in the text. Join them where the standard spelling needs one word.
  4. Check capital letters. Nouns are capitalised in German; verify names and technical terms in particular.
  5. Check separable verbs. German verbs split in a sentence, so the prefix can land far from the verb. Read these sentences twice.
  6. Export. Choose TXT, MD, DOCX, PDF, SRT or VTT for the next step.
What Potok can do
FeatureDetails
Where it worksTelegram bot @potok_now_bot; large files go through the bot's mini app
File sizeUp to 2 GB
ExportTXT, MD, DOCX, PDF, SRT, VTT
Speakers and timeSpeaker labels and timestamps can be switched on
LanguagesAbout 30 languages with good published results; a few more with warnings
TranslationThe transcript is not translated into another language
DictationA separate Potok desktop app lets you dictate text into any program; it recognises Russian speech

Common mistakes

Common mistakes and how to fix them
MistakeWhat happensFix
Leaving split compoundsThey look like typos and slow down reading.Join them during the edit.
Assuming Swiss or Austrian speech matches the testRegional speech differs from the standard form.Test a short sample first.
Quoting 4.3 as our resultIt comes from tests of others.Say where the number is from.
Sending a talk with many English termsMixed speech is harder to write down.Proofread those terms.

More

See the full list of languages, the tools and the glossary.

Frequently asked questions

How good is German?

The published word error rate is 4.3 on large-v3, among the best on our list, from tests of others.

Why the range 4.30 to 5.8?

Different published tests of the model report different values, so we show the lowest as the headline and the range as context.

Can I send a video?

Yes, a video is read like audio, within the 2 GB file limit of the bot.

Is the German translated?

No. The German text stays German, which is the language of the recording.

Can I get timestamps?

Yes, switch them on before sending, and export SRT or VTT.

Sources

Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.

  1. Word error rate (Wikipedia), Wikipedia · checked
  2. FLEURS speech dataset (Google, Hugging Face), Hugging Face · checked
  3. Canary-1B-v2 paper, Table 11 (Whisper large-v3 on FLEURS) · checked
  4. Whale paper, Table 5 (Whisper large-v3 on FLEURS) · checked
  5. 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