Hi there, Iām Yuxiang Zhao š
Welcome to my GitHub! My research interests focus on speech processing, including text-to-speech synthesis, synthetic speech deepfake detection and watermarking, and speech translation.
š¤ About Me
- Name: Yuxiang Zhao (čµµå®ēæ)
- Current Location: Suzhou, China
- Education:
- Master of Science in Computer Science, Shanghai Jiao Tong University (Sep 2024 - Present)
- Lab: X-LANCE Lab
- Advisor: Prof. Xie Chen
- Bachelor of Engineering in Civil Engineering, Shanghai Jiao Tong University (Sep 2020 - Jun 2024)
- Internship: AISpeech (ęåæ
é©°), July 2025 - Present
š Projects
Cascaded Speech Translation System
- Duration: October 2025 - Present
- Description: Developed a cascaded speech translation system that enables real-time translation between different languages. The system combines automatic speech recognition, machine translation, and text-to-speech synthesis to provide seamless translation experience.
- Demo: https://translate.sjtuxlance.com/
Traceable TTS: Toward Watermark-Free TTS with Strong Traceability
- Duration: November 2024 - April 2025
- Description: Proposed a novel framework for model attribution in TTS systems. Instead of embedding watermarks, we train the TTS model and discriminator using a joint training method that significantly improves traceability generalization while preserving audio quality.
- Paper: https://arxiv.org/abs/2507.03887
- Code: https://github.com/zhaoyx239/Traceable-TTS
Research on Landslide Prediction in Sanxia Reservoir Area using Deep Learning Algorithms
- Duration: October 2022 - October 2023
- Description: Developed and implemented a landslide time prediction system using deep learning techniques. The project focuses on using U-net to predict landslide occurrences in Sanxia Reservoir Area, with an emphasis on enhancing prediction accuracy and computational efficiency.
š« Connect with Me
Thank you for visiting my profile! Feel free to reach out or explore my repositories.