Cross-Modal BERT for Text-Audio Sentiment Analysis

Benchmark Model Rank Results
multimodal-sentiment-analysis-on-mosiCM-BERT#4Accuracy: 84.5%F1 score: 84.5%