The house workforce discusses why it looks like everyone wants subtitles now, the AI that generates music from textual content, and a listing of open-source knowledge engineering tasks so that you can contribute to.
Episode notes:
It’s not simply you: All of us want subtitles now.
Google introduces MusicLM, a mannequin that generates music from textual content. The examples are pretty-mind blowing and lift massive questions on licensing and copyrights for non-AI creators.
Taking the uncanny valley to a brand new low? Nvidia’s streaming software program now contains a characteristic that deepfakes eye contact.
Beware the possibly harmful intersection of AI and stan Twitter.
Due to Siavash Kayal, a fan of the present and knowledge engineer at Cleo, who despatched alongside an excellent listing of open-source knowledge engineering tasks of us can work on.
Right now we’re shouting out Stellar Query badge winner Paragon for asking Open two situations of a file in a single Visible Studio session.
The put up The AI that writes music from textual content (Ep. 535) appeared first on Stack Overflow Weblog.