Seyed Omid Sadjadi

Papers

Total Citations

H-Index

About

Seyed Omid Sadjadi is a researcher whose work sits at the compelling intersection of speech processing, affective computing, and multimodal machine learning. His most recognized contributions center on automatic emotion recognition, where he has pioneered the application of transfer learning techniques to bridge speaker recognition and natural language understanding. Most notably, his work on multimodal emotion recognition leverages pretrained speaker recognition models alongside BERT-based language representations, demonstrating that knowledge embedded in one domain can be powerfully repurposed to decode human emotional states — a challenge with far-reaching implications for human-computer interaction. With over 60 citations on this line of research alone, Sadjadi's contributions have resonated strongly within the speech and affective computing communities. His work addresses real-world applications including call center analytics, gaming, and intelligent personal assistants — contexts where understanding emotional nuance can meaningfully transform user experience. By combining acoustic and linguistic modalities through transfer learning, he has helped establish a more robust framework for emotion recognition that moves beyond single-channel approaches. His research reflects a broader commitment to making artificial intelligence more emotionally aware, positioning him as a meaningful contributor to the next generation of intelligent, human-centered systems.