BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Text-and-audio methods - Cătălina Cangea\, ex-Google DeepMind
DTSTART:20250204T130000Z
DTEND:20250204T140000Z
UID:TALK226306@talks.cam.ac.uk
CONTACT:Mateja Jamnik
DESCRIPTION:This talk supports the R255 Advanced Topics in Machine Learnin
 g module on Multimodal Learning and provides a bird’s eye view of the ra
 pidly evolving text-audio landscape\, with a focus on music as a primary e
 xample of audio data. I will first present types of tasks that exist in th
 is space\, then discuss data curation challenges and follow with an overvi
 ew of some existing retrieval and generation methods\, including a quick p
 rimer on diffusion models. Finally\, I will describe current evaluation me
 trics and their limitations.\n\n"You can also join us on Zoom":https://cam
 -ac-uk.zoom.us/j/83400335522?pwd=LkjYvMOvVpMbabOV1MVTm8QU6DrGN7.1
LOCATION:Lecture Theatre 2\, Computer Laboratory\, William Gates Building
END:VEVENT
END:VCALENDAR