The code for our INTERSPEECH 2020 paper - Jointly Fine-Tuning "BERT-like'" Self Supervised Models to Improve Multimodal Speech Emotion Recognition