Meta Spirit LM: Open multimodal language model that freely mixes text and speech

Posted by anjneymidha on Friday October 18, 2024 @05:28PM
from the twitter.com dept.

( 4 comments | 12 points | on HN )

Re:Meta Spirit LM: Open multimodal language model that freel... by anjneymidha on Friday October 18, 2024 @05:28PM (#41883656)

[ Parent | on HN ]

Re:Meta Spirit LM: Open multimodal language model that freel... by elevenoh4 on Friday October 18, 2024 @06:24PM (#41884041)

anyone hosting? want to try it

[ Parent | on HN ]

Re:Meta Spirit LM: Open multimodal language model that freel... by lostmsu on Friday October 18, 2024 @06:10PM (#41883962)

The generated little llama story here https://speechbot.github.io/spiritlm/ has some issues, and appears to be less coherent than 3.1-8B

[ Parent | on HN ]

Re:Meta Spirit LM: Open multimodal language model that freel... by lostmsu on Friday October 18, 2024 @06:06PM (#41883929)

Says non-commercial when requesting access

[ Parent | on HN ]