voicebox-pytorch icon indicating copy to clipboard operation
voicebox-pytorch copied to clipboard

Published 20 hours ago • verified publisher icon lucidrains

speartts model

Open happy-machine opened this issue 1 year ago • 3 comments

Hi there

Ive tried building and saving the model using the speartts repo but the structure of the output doesnt align. Have also tried matching the parameters in both.

Is it possible to get more information on what text_to_semantic.load('spear/speartest.pt') expects or is there a trained model i can download?

Thanks!

happy-machine avatar Nov 12 '23 13:11 happy-machine

the same question

Jon-Zbw avatar Dec 29 '23 01:12 Jon-Zbw

Hey @happy-machine.

You can train your own model from this repo: https://github.com/lucidrains/spear-tts-pytorch.

As far as I know, there's no pretrained models for this, or atleast I didn't find any

wassimseif avatar Jan 18 '24 11:01 wassimseif

Thanks that’s really helpful!DanDisclaimer: I often send emails at times that suit my family and work commitments but I don’t expect a reply out of office hours. On 18 Jan 2024, at 18:33, Wassim Seifeddine @.***> wrote: Hey @happy-machine. You can train your own model from this repo: https://github.com/lucidrains/spear-tts-pytorch. As far as I know, there's no pretrained models for this, or atleast I didn't find any

—Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: @.***>

happy-machine avatar Jan 18 '24 11:01 happy-machine