A four-minute voice message is a small thing to send and a large thing to receive. You cannot skim it, you cannot search it, and you cannot listen to it in a meeting, on a train without headphones or next to a sleeping child. The person who sent it knows what they said; you have to play it, possibly twice, to find the one address or date that mattered. Turning it into text removes all of that, and it also makes the content usable later, when you need to quote the plumber's price or the exact time your friend suggested.
WhatsApp stores voice notes as Opus audio inside an OGG container, with the .opus extension. It is a format made for speech and it transcribes well despite its small size. Transcrify accepts the file as it is, so there is no need to convert it to MP3 first, and the language is detected automatically, which matters for the many families and teams that switch between two languages inside a single conversation.
On iPhone and Android, press and hold the voice message until the menu appears, then choose Forward or Share. From there you can save it to the Files app or Google Drive, send it to yourself by email, or use AirDrop to a Mac. The saved file is the original .opus, not a re-recording, so nothing is lost on the way. On WhatsApp Desktop and WhatsApp Web, hover over the message, open the small arrow menu and pick Download; the file lands in your usual downloads folder.
If you need several notes from one conversation, select them together in the chat and forward them to your own email in one go. Each attachment is a separate file, and each one is uploaded and transcribed separately here. For a very long exchange it is often quicker to export the whole chat with media from the chat settings, then pick the audio files from the export folder.
Voice notes are among the most personal recordings there are: a doctor explaining results, a partner, a child, a colleague speaking frankly because they thought nobody would ever read it. That is why where the transcription happens matters more here than for a podcast. Transcrify runs its own speech models on its own servers. The audio you upload is not sent to a third-party speech API, and the same applies to the free transcriptions you make without an account.
Recent versions of WhatsApp include a transcription feature of their own. It runs on the phone, which is good for privacy, but it covers only a handful of languages, it works only on messages received after the feature was switched on, and it is not available on every device. This page is for everything that falls outside those limits: notes you saved months ago, languages the app does not handle, files received on a desktop, and conversations where you want a copy of the text outside the app.
As a guest you can read each transcript and copy the text. A free account keeps a history of everything you have transcribed, lets you correct a misheard name, and allows downloads as TXT or as timed subtitles. It comes with 5 hours across 10 files, which is a great many voice notes, and the transcripts you made here as a guest are moved into the account when you sign up. AI chat, summaries and export to PDF or Word are part of the paid plans.