Vivek Govindan

Title: Innovations by Vivek Govindan in Speech-to-Text Technology

Introduction

Vivek Govindan is an accomplished inventor based in Redmond, WA (US). He has made significant contributions to the field of speech-to-text technology, holding 2 patents that enhance the accuracy and efficiency of automatic speech recognition systems. His work is particularly focused on improving the transcription of rare and out-of-vocabulary words.

Latest Patents

One of Vivek's latest patents is titled "Personalized batch and streaming speech-to-text transcription of audio." This innovation introduces techniques that reduce the error rate of automatic speech recognition (ASR) systems when transcribing rare and out-of-vocabulary words. The methods achieve personalization of connectionist temporal classification (CT) models through adaptive boosting, which biases predictions at the sub-word level. This approach allows for improved accuracy in transcribing custom vocabularies without the need to retrain the ASR system.

Another notable patent is the "Model training system for custom speech-to-text models." This invention outlines a transcription service that can create custom speech-to-text models tailored for specific domains. The service receives training data and instructions, determining schedules for executing tasks based on dependencies. The trained models can then be deployed as part of a network-accessible service, enabling users to convert audio into text effectively.

Career Highlights

Vivek Govindan is currently employed at Amazon Technologies, Inc., where he continues to innovate in the realm of speech recognition technology. His work has been instrumental in advancing the capabilities of ASR systems, making them more adaptable to various user needs.

Collaborations

Vivek collaborates with talented individuals such as Sravan Babu Bodapati and Katrin Kirchhoff, contributing to a dynamic and innovative work environment.

Conclusion

Vivek Govindan's contributions to speech-to-text technology exemplify the impact of innovation in enhancing communication tools. His patents not only improve transcription accuracy but also pave the way for more personalized and efficient speech recognition systems.