Index-MSR: A high-efficiency multimodal fusion framework for speech recognition

Jinming Chen, Lu Wang, Zheshu Song +1

Driven by large scale datasets and LLM based architectures, automatic speech recognition (ASR) systems have achieved remarkable improvements in accuracy. However, challenges persis…