Location History:
- Jiangsu, CN (2023)
- Nanjing, CN (2024)
Company Filing History:
Years Active: 2023-2024
Title: Innovations of Zhiqiang Mao in Voice Conversion and Speech Synthesis
Introduction
Zhiqiang Mao is a prominent inventor based in Jiangsu, China. He has made significant contributions to the fields of voice conversion and speech synthesis, holding three patents that showcase his innovative approaches to these technologies.
Latest Patents
One of his latest patents is a "Voice conversion system and training method therefor." This invention proposes a speech conversion scheme for non-parallel corpus training, eliminating the dependence on parallel text. It addresses the technical challenges of achieving speech conversion under limited resources and equipment. The system includes a trained speaker-independent automatic speech recognition model that can be utilized for any source speaker. This innovation improves pronunciation accuracy and reduces recognition errors in automatic speech recognition systems.
Another notable patent is the "GAN-based speech synthesis model and training method." This invention provides a method for converting text into audio using a GAN-based approach. The process involves obtaining to-be-converted text, converting it into phonemes, and digitizing the data to generate a text vector for input into the speech synthesis model. The accuracy of the generated audio is significantly enhanced through adversarial training between the generator and discriminator, leading to improved audio quality.
Career Highlights
Zhiqiang Mao is currently employed at Nanjing Silicon Intelligence Technology Co., Ltd. His work at this company has allowed him to further develop his innovative ideas and contribute to advancements in speech technology.
Collaborations
He collaborates with talented coworkers, including Huapeng Sima and Xuefei Gong, who share his passion for innovation in the field of speech technology.
Conclusion
Zhiqiang Mao's contributions to voice conversion and speech synthesis demonstrate his commitment to advancing technology in these areas. His patents reflect a deep understanding of the challenges in speech processing and offer innovative solutions that enhance audio quality and usability.