Speech and Language AI

What We Do

Welcome to the Speech, Language, Perception, and Reasoning (SLPR) Lab. Our research focuses on speech and language AI, aiming to enhance an AI system's ability to perceive and reason. We explore the entire process - from training and optimizing AI models to implementing them efficiently - while striving to develop practical technologies for real-world applications.

Research topics include (but are not limited to):

• Speech and language multimodal generative AI

• Efficient deep learning model design and algorithm development

• Real-time speech and language processing on-device

What We Value

The four words in our lab's name reflect the core areas we pursue:

1. Speech is the foundation of language processing (Speech to Language)

2. Language shapes our understanding of the world (Language to Perception)

3. Accurate Perception enables logical reasoning (Perception to Reasoning)

4. Advanced Reasoning further enhances speech and language capabilities (Reasoning to Speech & Language)

We believe these four elements are closely interconnected, and that combining them can bring us closer to developing practical and efficient AI and solving important problems. As AI gradually takes on more tasks traditionally done by humans, the values that truly stand out will be the ability to speak clearly, write logically, perceive accurately, and reason rigorously. Our lab aims to cultivate these strengths in both our research and our members.