Upload a .wav or .mp3 file:
Longer recordings (of at least 30 minutes) are recommended. Your uploaded files are stored temporarily on the Dartmouth servers in order to process your job, and deleted after.

or copy and paste a link to a YouTube video:
Long, single-speaker videos with no music work best.

Dialect of the speaker: Standard American Southern
Selecting the closest appropriate dialect for the acoustic model may increase transcription accuracy. Other dialects may be added in the future.

Speech Type Free speech or reading passage Word list
If your recording contains both styles, select the free speech option.

Your e-mail address:
We will not store or distribute your address.

submit