Voice messages, as text.
VM-Text is a Discord app that transcribes voice messages. Right-click one, pick Apps, and a few seconds later you have the words. The audio is never saved.
Free. Works in DMs, group chats and servers.
What it does
Transcribe (private)
Only you see the result. Good for DMs, meetings, or anywhere you can't play audio.
Transcribe (public)
Posts the transcript in the channel so everyone there can read it.
Automatic replies
In a server, the bot can reply to every voice message by itself. Anyone with Manage Server can switch that off.
A queue
Requests are handled one at a time, in order. If you have to wait, it tells you your place.
Done once
Each voice message is transcribed a single time. Ask again later and you get the saved text straight away.
Languages
It works out the language on its own. If it keeps guessing wrong, pick one from the list of 21 in /config.
Settings
Run /config to choose your language and whether replies are embeds or plain text. Server admins can set a default language and the automatic replies.
Using it
- Add VM-Text to your Discord account, or to a server.
- Right-click a voice message (long-press on mobile) and open Apps, then VM-Text.
- Choose private or public. The transcript turns up in a few seconds.
Transcribe (public)VM-TextYour data
Voice messages are transcribed on my own server with Whisper, an open-source model. The audio isn't sent to any outside service, and it's thrown away as soon as the text exists.
What is kept:
- the transcript text, so the same message isn't transcribed twice (it's deleted if the message is deleted);
- your settings: language, output style, and the automatic replies setting for a server.
No ads, no tracking, no analytics. The full details are in the Privacy Policy and the Terms.
Questions
Is it free?
Yes. It's a personal project with no ads and nothing to pay for.
Does it keep my voice messages?
No. The audio is held in memory while it's transcribed and then discarded. The transcript text is saved so each message is only done once, and it's deleted if the message is deleted.
Why does it ask for the Message Content permission?
So it can tell that a message in a server is a voice message and fetch the audio. It doesn't read or store ordinary text messages.
Account or server, which should I pick?
On your account you can use the Transcribe commands in any DM, group chat or server, but only when you ask. Added to a server, the bot can also reply to every voice message there automatically. You can have both.
How do I turn off the automatic replies?
Someone with the Manage Server permission can run /config and switch automatic transcription off.
The transcript is wrong.
Background noise, strong accents, names and people talking over each other all cause mistakes. Setting the right language in /config usually helps.
Who runs this?
Lavery.Jonez, in Scotland. You can reach me at contact@laveryjonez.uk.
Give it a go on your own account first.
It takes a couple of clicks, and you can remove it just as easily.