This inventor holds 5 USPTO granted patents and 4 published patent applications and 3 EPO patents. Top assignee: Google Inc.. Active years: 2025-2026.
Company Filing History:
Years Active: 2025-2026
Title: Innovations of Shaojin Ding in Speech Recognition Technology
Introduction
Shaojin Ding is a prominent inventor based in Mountain View, CA, known for his contributions to the field of speech recognition technology. With a total of three patents to his name, Ding has made significant advancements that enhance the capabilities of automated speech recognition systems.
Latest Patents
One of his latest patents is the "Unified cascaded encoder ASR model for dynamic model sizes." This innovative automated speech recognition (ASR) model features a first encoder that processes a sequence of acoustic frames, generating higher-order feature representations. The model includes a second encoder and decoder, which work together to produce probability distributions over possible speech recognition hypotheses.
Another notable patent is the "4-bit conformer with accurate quantization training for speech recognition." This method involves training an ASR model using quantization-aware techniques with native integer operations. The trained model is quantized to a fixed-bit width, ensuring efficient performance on user devices.
Career Highlights
Shaojin Ding is currently employed at Google Inc., where he continues to push the boundaries of speech recognition technology. His work has been instrumental in developing models that improve the accuracy and efficiency of speech recognition systems.
Collaborations
Ding collaborates with talented colleagues, including Qiao Liang and Yanzhang He, who contribute to the innovative projects at Google Inc. Their teamwork fosters an environment of creativity and technological advancement.
Conclusion
Shaojin Ding's contributions to speech recognition technology through his patents and work at Google Inc. highlight his role as a leading inventor in this field. His innovative approaches continue to shape the future of automated speech recognition systems.
