A January 13, 2026 publication presents Pocket TTS, a 100-million-parameter text-to-speech model described as running on a laptop without a GPU and offering voice cloning.
In a publication dated January 13, 2026, Kyutai Labs introduced Pocket TTS, a text-to-speech model with 100 million parameters. According to the post, it offers voice cloning and can run on a laptop without a GPU; the publication describes it as open-source.
The announcement points to a self-hosted TTS option for engineers to evaluate, but the supplied material includes no benchmark results or performance details. To verify the claims and terms of use, consult the original publication and the project's technical materials. If you use AI to study them, apply your organization's policy to personal or confidential data before processing it.