You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I am working on benchmarking audio models for Qwen2Audio and would really love to see what features the model pays attention to for tasks like emotion recognition. would this be a feasible feature to add?
Hi!
I am working on benchmarking audio models for Qwen2Audio and would really love to see what features the model pays attention to for tasks like emotion recognition. would this be a feasible feature to add?