Joel Eric Shor

Title: Innovations of Joel Eric Shor

Introduction

Joel Eric Shor is a prominent inventor based in Mountain View, California. He has made significant contributions to the field of speech technology, holding a total of 8 patents. His work focuses on advancing methods for detecting synthetic speech and improving speech classification systems.

Latest Patents

One of his latest patents is titled "Self-supervised speech representations for fake audio detection." This method involves receiving audio data that characterizes speech and generating audio feature vectors using a trained self-supervised model. It also includes the use of a shallow discriminator model to score the presence of synthetic speech in the audio data. If the score meets a certain threshold, the method determines that the speech comprises synthetic speech.

Another notable patent is "Methods and systems for implementing on-device non-semantic representation fine-tuning for speech classification." This innovation allows a computing system to obtain audio data with a speech portion and train a neural network to learn a non-semantic speech representation. The system evaluates the performance of this representation and fine-tunes it based on downstream tasks, ultimately generating a model that operates locally on mobile devices.

Career Highlights

Joel Eric Shor is currently employed at Google Inc., where he continues to push the boundaries of speech technology. His work has been instrumental in developing systems that enhance the accuracy and reliability of speech recognition and classification.

Collaborations

Throughout his career, Joel has collaborated with notable colleagues, including George Dan Toderici and Sung Jin Hwang. These collaborations have contributed to the advancement of innovative technologies in the field.

Conclusion

Joel Eric Shor's contributions to speech technology through his patents and work at Google Inc. highlight his role as a leading inventor in this domain. His innovations are paving the way for more sophisticated speech recognition systems and applications.