Shiliang Zhang

Title: Shiliang Zhang: Innovator in Speech Recognition Technology

Introduction

Shiliang Zhang is a prominent inventor based in Hangzhou, China, known for his significant contributions to the field of speech recognition technology. With a total of seven patents to his name, Zhang has developed innovative methods and systems that enhance the accuracy and efficiency of speech processing.

Latest Patents

Zhang's latest patents include a "Streaming end-to-end speech recognition method, apparatus and electronic device." This invention describes a method that extracts and encodes speech acoustic features from a voice stream in frame units. It involves block processing to predict activation points that need encoding and outputting. The decoder then decodes these points to produce a recognition result, improving the system's robustness against noise and enhancing performance and accuracy.

Another notable patent is the "Speech processing method, speech encoder, speech decoder and speech recognition system." This method processes a speech signal using two neural networks, where the first network offers higher computational efficiency, while the second provides greater accuracy. The target feature information representing semantics in the speech signal is determined based on the outputs of both networks.

Career Highlights

Throughout his career, Shiliang Zhang has worked with leading companies in the technology sector, including Alibaba Group Holding Limited and Alibaba Damo Technology Co., Ltd. His experience in these organizations has allowed him to collaborate on cutting-edge projects that push the boundaries of speech recognition technology.

Collaborations

Zhang has worked alongside notable colleagues such as Ming Lei and Wei Li, contributing to a collaborative environment that fosters innovation and creativity in technology development.

Conclusion

Shiliang Zhang's work in speech recognition technology exemplifies the impact of innovative thinking in enhancing communication systems. His patents reflect a commitment to improving the accuracy and efficiency of speech processing, making significant strides in the field.