←Back to NewsAI News/AudionewsAudioOpen SourceKyutai Opens Pocket TTS Training So Anyone Can Build CPU Voice CloningKyutai released the full training stack for its 100M-parameter Pocket TTS, letting anyone train a voice model from scratch for under $200.SourcekyutaiPublishedAug 25, 2026, 2:14 PMAuthorAlphaSignal NewsroomRead1 min readKyutai released the full training stack for its 100M-parameter Pocket TTS, letting anyone train a voice model from scratch for under $200.Reporting is indexed from AlphaSignal. Rights remain with the original publisher and cited sources.Read original report ↗Next readsTencent Hy · newsTencent's AuK Replaces 16 Speech Tools With One Open-Source ModelAlphaSignal · modelYuE2-3B Beats Suno Generating Editable Full Songs on a Single GPUAlphaSignal · repoaudio.cpp Runs Speech and Music AI up to 5x Faster Without Python
