WAV files from studio recorders, field recorders and dictation devices give the best transcription quality. Upload yours and get the text in minutes.
WAV is uncompressed, so no detail of the voice is lost. Recordings from devices like the Zoom H1n, Tascam or a USB microphone usually produce the most accurate transcripts.
WAV files are big, about 10 MB per minute in CD quality. If yours is over the upload limit, convert it to MP3 or M4A first; transcription quality stays almost the same.
Drop a video or audio file, or paste a public link. Choose the spoken language or let us detect it.
Our speech recognition turns speech into punctuated text with timestamps. Most files finish in minutes.
Copy the text, or download TXT, SRT and VTT subtitles. Your media file is deleted automatically.
No. Any sample rate from 8 kHz to 96 kHz works; audio is converted to 16 kHz internally, which is all speech recognition needs.
Yes. Channels are mixed to mono before transcription.
Yes. Speak punctuation naturally and the model adds commas and full stops for you.