What works
A public Reel, one that opens in a browser without signing in to Instagram, usually works. In our test on 23 October 2026, two public Reels with speech downloaded their sound in a second or two, and the text was written on the PC as with any YouTube video. The link looks like instagram.com/reel/….
Behind the link VoiceKeyboard uses yt-dlp, a free open-source downloader that knows Instagram, YouTube and many other sites. Whisper then turns the sound into text on your computer. Nothing is sent to a transcription service.
What does not work
Social networks and video sites such as YouTube, Instagram and Facebook keep fighting the downloading of their content, so this feature is not guaranteed to work: a link that works today may stop working after a change on their side.
- Anything that needs a login. Private accounts, close friends, and many ordinary posts: Instagram answers such requests with an empty page and asks to sign in. VoiceKeyboard does not use your Instagram account or your browser's cookies, so these links fail. In our test a regular post (
instagram.com/p/…) failed this way while Reels worked. - Reels without a voice. A clip with only music has nothing to transcribe, and a clip with no sound track at all ends with an error.
- Stories. Instagram shows them only to signed-in users, so they fail the same way.
When a site refuses a link, VoiceKeyboard checks for a newer yt-dlp and tries once more. If the link still fails, the message reads "YouTube refused the download" even for an Instagram link: it covers any site. Try again later, or check that the Reel opens in a private browser window without signing in.
Step by step
- In Instagram on the web, open the Reel and copy the address from the browser bar. On a phone, use Share → Copy link and send it to your PC.
- In VoiceKeyboard, open FILES and choose LINK.
- Press Paste.
- Under TRANSCRIPT DESTINATION, press Save as... and choose where the .txt goes.
- Leave LANGUAGE on Auto, or pick the language of the Reel.
- Press ▶ START.
The first time, the app offers to download yt-dlp (about 17 MB), FFmpeg (about 73 MB) and a Whisper model, and waits for your consent. After that a Reel takes as long as your PC needs for a minute of sound; an NVIDIA graphics card makes it several times faster. The video link guide explains the status line and the queue in detail.
Timestamps and several Reels
On Pro, tick "Timestamps in the text" and each phrase starts with its time, like [00:00:42]. In a 30-second Reel that matters less than in a lecture, so the bigger difference for Instagram is the queue: Pro takes a list of links, one per line, and writes a separate .txt for each Reel. That is how you collect the text of a whole series of tips or recipes in one go.
Free gives one link a day. A Reel is short, so the 60-minute limit never matters here; the count does. A link that fails does not use up the day's allowance.
Instagram captions or a transcript
| Captions in Instagram | VoiceKeyboard | |
|---|---|---|
| Where you see the text | on the video, while it plays | in a .txt file on your PC |
| A text file to keep, search and quote | no | yes |
| Times of phrases in the text | not stated | with Pro |
| Private and login-only posts | if you can see them | no |
| Needs installing | nothing | Windows app |
Languages and translation
Whisper hears about 99 languages and writes the words in the language they are spoken. It does not translate. If you need the Reel in another language, transcribe it first, then give the text to a translator or an AI chat.
Long videos follow the same route; for YouTube, with notes on timestamps, see how to get a YouTube transcript with timestamps. Every button of FILES is described in the help page Transcribing files and video links.
