How can I implement late fusion using audio and video features ?

I am trying to build a joint classifier for Bimodal Sentiment Analysis which takes two modalities(audio and video files) as inputs. Any suggestions, how can I concatenate the below audio and video...

07 August 2020 955 2 View