Jinsong Han

Title: Jinsong Han - Innovating Multimodal Speech Recognition in Hangzhou

Introduction

Jinsong Han is an accomplished inventor based in Hangzhou, China. He has made significant contributions to the field of speech recognition, particularly through his innovative approaches that integrate multiple modalities of input data.

Latest Patents

Han currently holds a patent for a "Multimodal speech recognition method and system, and computer-readable storage medium." This invention provides a sophisticated method for achieving high-accuracy speech recognition by incorporating both millimeter-wave signals and audio signals that contain speech information corresponding to a specific target user. The method includes calculating several coefficients and utilizing a fusion network that incorporates calibration and mapping modules to create a comprehensive feature set for improved speech recognition results.

Career Highlights

Jinsong Han is primarily associated with Zhejiang University, where his research and development efforts focus on advancing technologies that enhance the efficiency and accuracy of speech recognition systems. His work stands out in an era where accurate voice recognition is increasingly important in various applications, from assistive technologies to smart devices.

Collaborations

Throughout his career, Han has collaborated with notable colleagues such as Feng Lin and Tiantian Liu. These partnerships have contributed to the development and refinement of his innovative speech recognition methods, facilitating significant advancements in the field.

Conclusion

Jinsong Han's work in multimodal speech recognition not only highlights his talents as an inventor but also showcases the potential of integrating various technologies to improve user experience. His contributions at Zhejiang University and his active collaborations continue to pave the way for future innovations in the realm of speech technology.