Zhifeng Chen

USPTO Granted Patents = 39 (Patent Applications = 21)

EPO Patents = 6 (Patent Applications = 8)

Average Co-Inventor Count = 8.7

ph-index = 5

Forward Citations = 56(Granted Patents)

Location History:

Warrington, PA (US) (2018 - 2024)
Sunnyvale, CA (US) (2020 - 2024)

Company Filing History:

Merck Sharp + Dohme Corp.

Years Active: 2018-2025

Loading Chart...

39 patents (USPTO):Sort by Forward Citations

Abstract keywords

1. 12443806 - Machine translation using neural network models (2025-10-14),

2. 12371478 - Antibody neutralizing human respiratory syncytial virus (2025-07-29),

3. 12254865 - Multi-dialect and multilingual speech recognition (2025-03-18),

4. 12190860 - End-to-end text-to-speech conversion (2025-01-07),

5. 12175963 - Synthesis of speech from text in a voice of a target speaker using neural networks (2024-12-24),

6. 12148444 - Synthesizing speech from text using neural networks (2024-11-19),

7. 12106749 - Speech recognition with sequence-to-sequence models (2024-10-01),

8. 12106528 - Generating scene flow labels for point clouds using object labels (2024-10-01),

9. 12087273 - Multilingual speech synthesis and cross-language voice cloning (2024-09-10),

10. 12032920 - Direct speech-to-speech translation via machine learning (2024-07-09),

11. 11981726 - Antibody neutralizing human respiratory syncytial virus (2024-05-14),

12. 11922932 - Minimum word error rate training for attention-based sequence-to-sequence models (2024-03-05),

13. 11900915 - Multi-dialect and multilingual speech recognition (2024-02-13),

14. 11862142 - End-to-end text-to-speech conversion (2024-01-02),

15. 11848002 - Synthesis of speech from text in a voice of a target speaker using neural networks (2023-12-19),

Title: Innovations of Zhifeng Chen in Speech Synthesis Technology

Introduction

Zhifeng Chen, based in Sunnyvale, California, is a prominent inventor known for his groundbreaking contributions to speech synthesis and text-to-speech conversion technologies. With an impressive portfolio of 29 patents, his work has significantly impacted the field of artificial intelligence and natural language processing.

Latest Patents

Among his latest innovations, Zhifeng Chen has developed novel methods, systems, and apparatus for end-to-end text-to-speech conversion. These inventions encompass computer programs encoded on computer storage media that efficiently generate speech from text. One key system utilizes a sequence-to-sequence recurrent neural network that receives a sequence of characters in a chosen natural language and processes this sequence to produce a spectrogram of the verbal utterance.

Moreover, Chen's patent involves the synthesis of speech from text using the voice of a target speaker through advanced neural networks. This method includes obtaining an audio representation of a target speaker’s voice and generating a speaker vector through a trained encoder. The process results in realistic audio representations of the input text articulated in the target speaker’s voice, highlighting Chen’s commitment to enhancing the personalization of speech synthesis.

Career Highlights

Zhifeng Chen has held significant positions at leading technology companies, such as Google Inc. and Waymo LLC. Through his work, he has integrated innovative technologies into practical applications, showcasing his dedication to advancing the field of speech processing.

Collaborations

Throughout his career, Chen has collaborated with esteemed individuals in the industry, including Yonghui Wu and Ron J Weiss. Their joint efforts have propelled innovations in the realm of speech synthesis and contributed to the development of cutting-edge technologies.

Conclusion

Zhifeng Chen's contributions to the field of speech synthesis and text-to-speech conversion are a testament to his innovative spirit and expertise. With numerous patents to his name, he continues to pave the way for advancements in artificial intelligence, making communication technologies more efficient and personalized.

This text is generated by artificial intelligence and may not be accurate.