Re:StyleTTS2 – open-source Eleven Labs quality Text To Speech
by modeless on Sunday November 19, 2023 @09:58PM
(#38341707)
Yes, unfortunately these models take a lot of VRAM. It may be possible to do an 8GB version but it will have to compromise on quality of voice recognition and the language model so it might not be a good experience.
[ Parent
| on HN ]
Re:StyleTTS2 – open-source Eleven Labs quality Text To Speech
by joshspankit on Sunday November 19, 2023 @10:15PM
(#38341821)
This might be silly because of how few people it benefits, but could it be broken up on to multiple 8GB cards on the same system?
Re:StyleTTS2 – open-source Eleven Labs quality Text To Speech
by modeless on Sunday November 19, 2023 @10:37PM
(#38341987)
Yes, it absolutely could. You're right that this configuration is rare.
Re:StyleTTS2 – open-source Eleven Labs quality Text To Speech
by wahnfrieden on Monday November 20, 2023 @01:25AM
(#38343357)
The latest large models are 120B and 100k context such as Goliath and Tess XL