Enter inventor name

Winchester, MA, United States of America

John R Hershey

This inventor holds 45 USPTO granted patents and 10 published patent applications and 10 EPO patents, primarily in Speech Recognition. Top assignees: Mitsubishi Electric Research Laboratories, Inc., Microsoft Technology Licensing, LLC, Sony Corporation. Active years: 2007-2025.

USPTO Granted Patents = 45 (Patent Applications = 10)

% Patents Active = 60.0

EPO Patents = 10 (Patent Applications = 13)

Average Co-Inventor Count = 3.4

ph-index = 10

Forward Citations = 450(Granted Patents)

Location History:

La Jolla, CA (US) (2008 - 2009)
Seattle, WA (US) (2008 - 2009)
San Diego, CA (US) (2007 - 2010)
White Plains, NY (US) (2009 - 2014)
Winchester, MA (US) (2013 - 2023)

Company Filing History:

32 out of 1,694

Mitsubishi Electric Research Laboratories, Inc.

patents

6 out of 55,978

Microsoft Technology Licensing, LLC

University of California

International Business Machines Corporation

patents

7. 1 out of 1,422

Nuance Communications, Inc.

patents

where one patent can have more than one assignee

Years Active: 2007-2025

Loading Chart...

Areas of Expertise:

Speech Recognition

Audio Enhancement

Machine Learning

Signal Separation

Neural Networks

Multichannel Processing

Noise Reduction

Language Models

Contextual Understanding

Multimodal Fusion

Speech Signals

Acoustic Processing

45 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12217768 - Audio-visual separation of on-screen sounds based on machine learning models (2025-02-04),

2. 11756570 - Audio-visual separation of on-screen sounds based on machine learning models (2023-09-12),

3. 11133011 - System and method for multichannel end-to-end speech recognition (2021-09-28),

4. 11086918 - Method and system for multi-label classification (2021-08-10),

5. 10811000 - Methods and systems for recognizing simultaneous speech by multiple speakers (2020-10-20),

6. 10726856 - Methods and systems for enhancing audio signals corrupted by noise (2020-07-28),

7. 10672388 - Method and apparatus for open-vocabulary end-to-end speech recognition (2020-06-02),

8. 10593321 - Method and apparatus for multi-lingual end-to-end speech recognition (2020-03-17),

9. 10592800 - Neural networks for transforming signals (2020-03-17),

10. 10529349 - Methods and systems for end-to-end speech separation with unfolded iterative phase reconstruction (2020-01-07),

11. 10417498 - Method and system for multi-modal fusion model (2019-09-17),

12. 10176799 - Method and system for training language models to reduce recognition errors (2019-01-08),

13. 9881631 - Method for enhancing audio signal using phase information (2018-01-30),

14. 9842106 - Method and system for role dependent context sensitive spoken and textual language understanding with neural networks (2017-12-12),

15. 9837075 - Statistical voice dialog system and method (2017-12-05),

Title: John R. Hershey: Innovator in Audio-Visual Technology

Introduction

John R. Hershey, based in Winchester, MA, has made significant contributions to the field of audio-visual technology with an impressive portfolio of 34 patents. His innovative work focuses on the intersection of machine learning and audio source separation, offering advanced methodologies that enhance the auditory experience in multimedia applications.

Latest Patents

One of Hershey's latest patents includes the method for the audio-visual separation of on-screen sounds based on machine learning models. This apparatus and its associated methods optimize the separation of audio sources by processing an audio waveform linked with multiple video frames. By employing a neural network, the method estimates and generates audio embeddings that correspond to various audio sources. It also determines whether these audio sources match specific objects depicted within the corresponding video frames. This innovative approach ultimately allows for the prediction of an audio waveform that highlights the audio sources relevant to the on-screen objects, enhancing both clarity and user experience.

Career Highlights

Throughout his career, John R. Hershey has been associated with notable organizations such as Mitsubishi Electric Research Laboratories, Inc. and Google Inc. His tenure at these companies has allowed him to spearhead pioneering research and development projects that focus on machine learning applications in multimedia.

Collaborations

In his journey of innovation, Hershey has collaborated with esteemed colleagues including Jonathan Le Roux and Shinji Watanabe. These partnerships have fostered a dynamic exchange of ideas and expertise, significantly contributing to the advancements in audio-processing technologies.

Conclusion

John R. Hershey stands as a key figure in the realm of audio-visual technology innovation. His extensive patent portfolio and collaborative efforts underscore his commitment to enhancing audio clarity through cutting-edge methodologies. As technology continues to evolve, the impact of Hershey's work will undoubtedly play a critical role in shaping the future of multimedia experiences.

This text is generated by artificial intelligence and may not be accurate.

Please report any incorrect information to [email protected]