Umberto Cappellazzo

Continual learning for audio and speech processing. Parameter-efficient transfer learning of pre-trained models. Multimodal Large Language Models.