Free German Speech-to-Text

German is the most widely spoken native language in the European Union, with around 95 million native speakers. It is official in Germany, Austria, Switzerland, Liechtenstein, Luxembourg and parts of Belgium and Italy, which means German recordings turn up everywhere from Berlin newsrooms to Zurich boardrooms.

German also puts specific demands on a speech-recognition engine. Nouns are capitalised, so a transcript that gets this wrong is instantly recognisable as machine output — ConvertSpeech writes them correctly. Compound nouns are built by fusing words together rather than separating them, so a single spoken word can arrive as Rechtsschutzversicherung or Lebensmittelunverträglichkeit and must be written as one. And spoken numbers below a hundred run backwards: einundzwanzig is literally "one and twenty", which is why a naive system can turn 21 into 12.

Then there is the ß. It exists in Germany and Austria but not in Switzerland, where ss is used throughout — so the same sentence is spelled differently depending on where the speaker is from. Our engine follows the convention of the recording rather than forcing one standard, which is why the sample transcript below preserves the older spelling daß exactly as it was read.

Upload your audio or video and the engine writes out the spoken German as clean, searchable text with punctuation. It runs in your browser, needs no installation, and your first minutes are free with no account required.

This is a real 45-second recording of a human reader, run through our AI engine exactly as you would run your own file. Nothing was corrected afterwards. Press play and read along.

Worth noticing: the Russian names Abraham Petrowitsch Hannibal and Katharina come out correctly, the French rank en chef is kept in French rather than being germanised, and the historic spelling daß is preserved instead of being modernised to dass. Achtzehn and zweiundneunzigsten stay written out as spoken.

This is careful, clearly articulated reading — close to a lecture or a dictated note. A noisy café interview with three people talking over each other will not come out this clean; nothing automatic will. Recording quality remains the single biggest factor in the result.

Standard German (Hochdeutsch) is what most broadcast, business and academic recordings use, and ConvertSpeech is tuned for it. It handles the standard spoken across Germany, Austria and Switzerland reliably.

German also has strong regional variation — Bavarian and Austrian German in the south, Swiss German (Schweizerdeutsch) in Switzerland, and Low German in the north. Swiss German in particular differs enough from standard German that heavy dialect is challenging for any automatic system: it is closer to a group of Alemannic dialects than to an accent, with its own vocabulary and vowel system. If your recording is strongly dialectal, expect the transcript to render it in standard German spelling rather than reproducing the dialect.

Austrian German sits much closer to the standard and transcribes well, with occasional slips on distinctly Austrian vocabulary — Jänner for Januar, Erdäpfel for Kartoffeln, Sackerl for Tüte. These are worth a quick scan when you proofread.