Non Native

Research on non-native speech and language processing focuses on improving the accuracy and fairness of technologies interacting with non-native speakers. Current efforts concentrate on disentangling segmental and prosodic features in speech to better understand their impact on comprehension and perception, developing robust automatic speech recognition (ASR) systems using techniques like transfer learning and fine-tuning on diverse datasets including spontaneous learner speech, and mitigating biases in large language models (LLMs) and AI writing tools that disadvantage non-native English users. These advancements have significant implications for improving accessibility to technology and fostering more inclusive communication in diverse multilingual settings.