This inventor holds 140 USPTO granted patents and 33 published patent applications and 10 EPO patents, primarily in Acoustic Signal Processing (CPC class G10L15-063). Top assignees: Microsoft Technology Licensing, LLC, Tencent America LLC, Tencent Technology (Shenzhen) Company Limited. Active years: 2007-2026.
Inventors with similar research interests:
Location History:
- Kirkland, WA (US) (2007 - 2020)
- Bellevue, WA (US) (2022)
- Shenzhen, CN (2022)
- Bothell, WA (US) (2014 - 2024)
- Palo Alto, CA (US) (2023 - 2024)
Company Filing History:
67 out of 56,194 
63 out of 2,146 
9 out of 5,348 
Years Active: 2007-2026
Areas of Expertise:
Title: Dong Yu: Pioneer in Singing Voice Conversion and Multi-Channel Target Speech Separation
Introduction:
Dong Yu, a renowned inventor hailing from Kirkland, WA (US), has made impressive contributions to the field of speech processing and audio technology. With over 100 patents to his name, Dong Yu continues to push the boundaries of innovation as he seeks to enhance the way we understand and interact with sound. This article sheds light on some of his latest patents, notable collaborations, and his tenure at key companies in the industry.
Inventions and Patents:
Singing Voice Conversion:
One of Dong Yu's recent patents involves a groundbreaking method, computer program, and computer system for singing voice conversion. This invention allows for the seamless transformation of a first singing voice associated with one speaker to a second singing voice associated with another speaker. By encoding the context of specific phonemes and aligning them to target acoustic frames, Dong Yu's innovation generates mel-spectrogram features that enable the conversion of the original singing voice sample to a sample corresponding to the desired singing voice.
Multi-Modal Framework for Multi-Channel Target Speech Separation:
Another notable patent from Dong Yu showcases his expertise in multi-channel target speech separation. This invention revolves around a method, computer program, and computer system that successfully separate a target voice from among multiple speakers. By extracting video feature data from video content and combining it with audio data associated with each speaker, Dong Yu's framework effectively identifies and isolates the target voice from the surrounding voices.
Career Highlights:
Dong Yu has contributed his expertise to renowned companies in the technology industry, including Tencent America LLC and Tencent Technology (Shenzhen) Company Limited. These organizations are global leaders in various technological domains, and Dong Yu's involvement with them serves as a testament to his exceptional skills and knowledge.
Notable Collaborators:
Throughout his career, Dong Yu has worked closely with esteemed colleagues who share his passion for innovation. Two notable collaborators include Li Deng and Alejandro Acero. These individuals have been instrumental in advancing the field of speech and audio technology, and their collaborations with Dong Yu have undoubtedly led to cutting-edge inventions and solutions.
Conclusion:
Dong Yu's dedication and brilliance in the field of speech processing and audio technology have paved the way for exciting advancements. Through his patents, such as singing voice conversion and multi-channel target speech separation, Dong Yu has showcased his ability to push the boundaries of what is possible in the realm of sound technology. With his invaluable contributions and collaborations with industry giants, Dong Yu has undoubtedly earned his position as a pioneering innovator in the field.