CNN-RNN-Transformer Late-Fusion Network
Transformer/CNN/RNN fusion model analyzing both audio and lyrics of songs.
Developed a multi-modal neural network architecture leveraging Transformers, CNNs, and RNNs (LSTMs and GRUs) to jointly analyze the audio and textual aspects of songs. Incorporated and compared large language models (BERT, T5, GPT), using TensorFlow for training and fine-tuning. Improved accuracy by 34% over prior work, demonstrating the effectiveness of integrating multiple modalities.
