Jason Riesa

Title: Jason Riesa: Innovator in Speech Synthesis and Language Processing

Introduction

Jason Riesa is a notable inventor based in San Francisco, CA. He has made significant contributions to the fields of speech synthesis and language processing. With a total of 2 patents, Riesa is recognized for his innovative approaches to evaluating synthesized speech and identifying codemixed text.

Latest Patents

Riesa's latest patents include "Automated evaluation of synthesized speech using cross-modal and cross-lingual transfer of language encoding." This method evaluates synthesized speech by receiving a speech sample in a first language and determining a rating based on an artificial intelligence encoding model. The rating corresponds to the naturalness of the speech sample, utilizing a two-stage training process with diverse training data.

Another significant patent is "Identifying codemixed text." This method involves receiving codemixed text and segmenting it into tokens. Each token is analyzed to extract features, and a language identifier model predicts the probability distribution over possible languages for each token. The method assigns a language to each token through a greedy search based on the predicted probabilities.

Career Highlights

Jason Riesa is currently employed at Google Inc., where he continues to push the boundaries of technology in speech and language processing. His work has garnered attention for its innovative methodologies and practical applications in real-world scenarios.

Collaborations

Riesa collaborates with talented individuals such as Ankur Bapna and Daniel Gillick. Their combined expertise contributes to the advancement of projects within the realm of artificial intelligence and language technology.

Conclusion

Jason Riesa stands out as an influential inventor in the field of speech synthesis and language processing. His patents reflect a commitment to innovation and excellence in technology.