New Providence, NJ, United States of America

Sarangarajan Parthasarathy

This inventor holds 51 USPTO granted patents and 2 published patent applications, plus 1 CIPO patent and 3 EPO patents, primarily in Automatic Speech Recognition. Top assignees: At&t Intellectual Property I, LP, Microsoft Technology Licensing, LLC, At+t Corp.. Active years: 1999-2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Sarangarajan Parthasarathy. Retrieved from https://idiyas.com/inventor/sarangarajan-parthasarathy

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 51 (Patent Applications = 2)

% Patents Active = 66.7

CIPO Patents = 1

EPO Patents = 3 (Patent Applications = 7)

Average Co-Inventor Count = 2.8

ph-index = 16

Forward Citations = 885(Granted Patents)

Forward Citations (Not Self Cited) = 845(Dec 10, 2025)

Location History:

Watchung, NJ (US) (2012)
Sunnyvale, CA (US) (2019)
New Providence, NJ (US) (1999 - 2020)
Mountain View, CA (US) (2015 - 2022)

Company Filing History:

22 out of 16,777

At&t Intellectual Property I, LP

patents

17 out of 56,259

Microsoft Technology Licensing, LLC

Nuance Communications, Inc.

patents

Years Active: 1999-2026

Loading Chart...

Areas of Expertise:

Automatic Speech Recognition

Language Model Optimization

Transcription Correction

Voice Form-Filling

Intent Frame Suggestion

Pronunciation Learning

Speech Query Reformulation

Context-Aware Formatting

Token-Level Interpolation

User-Specific Recognition

Speaker Segmentation

Dialog Cue Processing

51 patents (USPTO):Sort by Forward CitationsExplore Patents

1. 12748800 - Low-resource, multi-lingual transformer models (2026-09-29),

2. 12354600 - Fast and efficient text only adaptation for factorized neural transducer (2025-07-08),

3. 11527238 - Internal language model for E2E models (2022-12-13),

4. 10607600 - System and method for mobile automatic speech recognition (2020-03-31),

5. 10529326 - Suggesting intent frame(s) for user request(s) (2020-01-07),

6. 10497367 - Flexible schema for language model customization (2019-12-03),

7. 10176219 - Interactive reformulation of speech queries (2019-01-08),

8. 9972311 - Language model optimization for in-domain application (2018-05-15),

9. 9947317 - Pronunciation learning through correction logs (2018-04-17),

10. 9892728 - System and method for mobile automatic speech recognition (2018-02-13),

11. 9824682 - System and method for robust access and entry to large structured data using voice form-filling (2017-11-21),

12. 9734826 - Token-level interpolation for class-based language models (2017-08-15),

13. 9672202 - Context-aware re-formating of an input (2017-06-06),

14. 9589562 - Pronunciation learning through correction logs (2017-03-07),

15. 9583104 - Suggesting intent frame(s) for user request(s) (2017-02-28),

Title: The Innovations of Sarangarajan Parthasarathy

Introduction

Sarangarajan Parthasarathy, based in New Providence, NJ, has made significant contributions to the field of automatic speech recognition with an impressive portfolio of 49 patents. His work has had a substantial impact on how machines understand human speech, delivering groundbreaking advancements in technology and user experience.

Latest Patents

Among Sarangarajan's most recent patents is the invention titled "Internal Language Model for E2E Models." This innovation involves a computer device designed to enhance automatic speech recognition by utilizing a trained end-to-end (E2E) model and an external language model, allowing for more accurate and efficient processing of speech features. The invention includes unique methodologies for computing integrated scores, providing a deeper insight into spoken input.

Another key patent is the "System and Method for Mobile Automatic Speech Recognition," which outlines a process for updating speech recognition parameters on mobile devices. This method involves the storage of user account-specific adaptation data and the dynamic generation of new adaptation parameters for improved accuracy in recognizing user utterances when communication channels are established.

Career Highlights

Sarangarajan has worked with prominent companies such as AT&T Intellectual Property II, L.P. and Microsoft Technology Licensing, LLC. His experience in these organizations has helped him refine his skills in developing innovative speech recognition technologies and pushing the boundaries of what's possible in this domain.

Collaborations

Throughout his career, Sarangarajan has collaborated with notable individuals including Richard Cameron Rose and Aaron Edward Rosenberg. These partnerships have enabled him to leverage diverse expertise and insights, enhancing the quality and impact of his inventions.

Conclusion

Sarangarajan Parthasarathy's contributions to the field of automatic speech recognition through his extensive array of patents are remarkable. His work not only showcases his inventiveness but also demonstrates a commitment to advancing technology that improves communication between humans and machines. With a promising career ahead, Sarangarajan continues to be a pivotal figure in innovation and technology.

Profile summary based on public USPTO records.