Company Filing History:
Years Active: 2020-2023
Title: Shubhahrata Sengupta: Innovator in Real-Time Neural Text-to-Speech Technology
Introduction
Shubhahrata Sengupta is a prominent inventor based in Menlo Park, CA (US). He has made significant contributions to the field of text-to-speech technology, particularly through his innovative use of deep neural networks. With a total of 2 patents to his name, Sengupta is recognized for his advancements in creating production-quality systems that enhance the efficiency and flexibility of text-to-speech applications.
Latest Patents
Sengupta's latest patents focus on real-time neural text-to-speech systems. These systems are constructed from deep neural networks and comprise five major building blocks: a segmentation model for locating phoneme boundaries, a grapheme-to-phoneme conversion model, a phoneme duration prediction model, a fundamental frequency prediction model, and an audio synthesis model. The segmentation model utilizes deep neural networks with Connectionist Temporal Classification (CTC) loss for phoneme boundary detection. Additionally, a variant of WaveNet was developed for the audio synthesis model, which requires fewer parameters and trains faster than the original. This innovative approach allows for simpler and more flexible systems compared to traditional text-to-speech systems, which often require extensive feature engineering and domain expertise. Inference with these systems can be performed faster than real time, showcasing their efficiency.
Career Highlights
Sengupta is currently employed at Baidu USA LLC, where he continues to push the boundaries of text-to-speech technology. His work has positioned him as a key player in the development of advanced neural network applications in this field.
Collaborations
Throughout his career, Sengupta has collaborated with notable colleagues, including Sercan Omer Arik and Mike Chrzanowski. These collaborations have further enriched his work and contributed to the advancements in text-to-speech technology.
Conclusion
Shubhahrata Sengupta is a leading inventor in the realm of neural text-to-speech systems, with a focus on creating efficient and flexible solutions. His contributions are paving the way for future innovations in this exciting field.