One clip, one transcript

Speech to text

Classic speech-to-text apps want a microphone or a document workflow. Droptext is narrower: a short file you already have, turned into text you can paste.

Start on the plate

Pick a language when you know it

Auto is fine for one clean speaker. If you know the language, set it. Mixed speech follows the stronger language.

Limits

Two minutes of speech. Longer files: trim the first two minutes on the plate.
About 200 MB for the file you drop.
No speaker labels, live mic, or SRT download.

Questions

How accurate is it?

Quality depends on the recording: a clear voice works better than talk under loud music.