Turn a long German recording into a full transcript
Upload a German audio or video file, from a two-hour Experteninterview to a day of lectures split across several files, and get every word written out with a timestamp at the start of each paragraph. A recording of up to 15 minutes is free in full. Anything longer shows its first part free, and the whole recording costs $5, paid once, for up to 6 hours.
- Umlauts and ß kept, nouns capitalized the way German spelling requires
- Swiss German speech comes out as standard written German, which is how most Swiss people write
- "Sie" and "du" stay exactly as the speakers used them
- Optional English translation under every German paragraph
- Up to 6 hours for $5; Unlimited covers up to 12 hours per job
Real output from a demo recording. Click a timestamp to see where it sits in the audio.
Interview transcripts for a Bachelor or Master thesis
A Leitfadeninterview or Experteninterview for an Abschlussarbeit has to be transcribed in full before you can code it. Upload the recording, check the free first part, and the whole interview comes back with a timestamp at each paragraph and a new paragraph when the speaker changes. Rename "Speaker 1" and "Speaker 2" to I and B, or to Interviewer and Befragte, with Find and replace on screen, then download Word for the appendix.
The tool writes what was said and does not apply a university transcription system. If your department wants pauses in brackets, marked emphasis or smoothed dialect, that is a step you do in Word afterwards. One more thing on long interviews: speaker numbers are assigned in blocks of about ten minutes, so the same person can be Speaker 1 early on and Speaker 2 an hour later. Read through once and fix the labels before you rename them globally.
Swiss German, Austrian German and dialect
Schweizerdeutsch is spoken, not usually written: Swiss people write standard German. The transcript follows that habit and renders a Zurich or Bern interview in standard German spelling. The meaning comes through, but dialect words with no standard equivalent can be approximated, written as heard or marked [inaudible], and the feel of the original is lost. If your research needs a dialect transcription, for example in Dieth spelling, this tool does not produce one. Swiss spelling writes ss where Germany writes ß; the transcript may follow either convention, so use Find and replace if your text must be consistent.
Austrian speakers are written in standard German too, and words that are standard in Austria, such as "Jänner" or "heuer", stay as said. Strong Viennese, Tyrolean, Bavarian or Swabian dialect makes mistakes more likely. Look at the free first part before you pay; it was transcribed exactly the way the rest will be.
Hearings, Anhörungen and official meetings
People bring recordings of asylum Anhörungen, works council meetings, Gemeinderat sessions and hearings they were permitted to record, often because a lawyer abroad or a relative needs an English version. Choose English under "Also translate" and each German paragraph is followed by its English translation. The result is a draft made by an AI speech model. It is not a certified transcript and not a sworn translation (beglaubigte Übersetzung); where a court or authority demands one, use this text to find the passages that matter and hand those to a sworn translator.
Before you upload
- Set the spoken language to German when the file opens with English small talk or music
- Name multi-part files 01, 02, 03 so the parts are joined in the right order
- Check technical terms, long compound words and the names of people and places
- Keep the original audio: the timestamps take you back to any sentence you want to quote
How the tool works, what it costs and how your files are handled is on the Long Audio & Video to Text page; prices for every tool are on the pricing page.
Last updated
Questions and answers
Can ChatGPT transcribe a 90-minute German interview?
For a short, clear clip a chat assistant often does fine. For a long file it tends to hit an upload limit or return a shortened version, and it cannot give reliable timestamps across the whole recording. This tool transcribes the file in parts of up to ten minutes and joins them in order, so every minute is written out.
Does it understand Schwiizerdütsch?
It writes Swiss German speech as standard German, which is how Swiss text is normally written. Dialect words may be approximated or marked [inaudible], so judge the free first part of your own recording.
What does a three-hour German lecture cost?
One payment of $5 covers any recording up to 6 hours. With several long recordings in one week, Unlimited at $19 a week covers up to 12 hours per job and can be cancelled any time.
Can my supervisor get an English version?
Yes. Choose English under "Also translate" before you upload, and the English appears under each German paragraph, with the German first.
Who can see the recording?
Only you. The file goes to an AI model provider solely to be transcribed and is not used to train models. Jobs started without an account are deleted after 7 days, and you can delete yours at any time.