How to transcribe audio to text on your computer
Three ways to turn a recording into text, and the steps to do it on your own computer with DijiFlow Dictate, where th...
You have %itemCount% in your cart.Total being %total%
How DijiFlow Dictate turns a video's soundtrack into timed subtitle lines on your computer, and which of its four outputs to pick.
Auto captions turn the speech in a video into timed subtitle lines, without typing a word of it. DijiFlow Dictate makes them on your own computer from a video or audio file on your drive, and gives you the result in one of four formats. Whether you need captions for videos you publish or a subtitle file for a client, the steps are the same.
The app transcribes the soundtrack, times every word, and groups the words into lines. Lines follow the speech and split at clauses, not in the middle of a phrase, within a line width set for each language, and the reading speed sets how long each line stays on screen. In the Advanced view, Reading speed offers Adult and Children.
This takes longer than a plain transcript, because every word is timed and every line is checked. After your first run, the app shows how long a new file will take on your computer.
The labels below are the ones the English Mac app shows.
For subtitles the app suggests Large, the largest model, because smaller models miss words in music and noise (Large is part of Pro and the trial). The subtitles page shows the four formats side by side.
One detail from the app itself: SRT and WebVTT are files a player loads beside the video, and QuickTime Player cannot open them. To watch the result on a Mac, pick one of the MP4 choices. The MP4 outputs also need a video your computer can read; if it cannot, the app asks you to convert the file to MP4 or MOV first.
In the Mac app, the free version makes subtitles for the first 3 minutes of each file, with a smaller model. It is free forever, with no account and no card, which is enough to see how the lines look on your own footage. Pro and the 30-day trial make subtitles for files of any length with the largest model. On Pro and the trial, turn on Identify Speakers in the Advanced view, and each new speaker's line starts with a hyphen.
Long files are fine: a 90-minute film stays at around 1 GB of memory while it runs. The whole job happens on your computer. Your audio and text are never uploaded.
Auto captions are a first draft. Names, numbers, technical terms and overlapping voices are where errors gather, and the app itself reminds you to check the result. Because SRT and WebVTT are plain text, you can correct a line in any text editor and keep the timing as it is. On Pro and the trial, Custom Terms in the Vocabulary tab help with names that come up again and again.
Set Output language to a language other than the one spoken, and the subtitles come back translated. Subtitles cover 29 languages, and the picker marks the weaker pairs as beta or reduced accuracy. Translated subtitles are part of Pro and the 30-day trial.
These captions are made from files. DijiFlow Dictate does not caption a live stream or a video call as it happens, and it does not take a link to a video on a website: the video has to be a file on your drive. As a subtitle generator, it writes new subtitles from the audio; it does not convert or edit a subtitle file you already have.
If you publish video regularly, the page for creators collects the parts of the app that matter for that work. To make your first subtitle file, download DijiFlow Dictate.