Deepfake.no

News · Analysis · Research · Intelligence

Search

Technology

New model clones voices from just three seconds of audio

An imagined research model shows how little audio is needed to imitate a voice.

By Deepfake.no-redaksjonen
9 Nov 20265 min read
Updated 12 Nov 2026Les på norsk

This article is a fictional example created for demonstration. People, events and figures are invented.

Short audio clips from social media may be enough to create a convincing copy of a voice.

What does it mean for you?

Agree on a code word with your family, and be sceptical of unexpected calls asking for money.

Sources

  1. Fiktiv demokilde – ingen ekte studie – Deepfake.no (eksempel)
ShareLinkedInXE-post