Zhengyang Li, M.Sc.

Contact Details

Zhengyang Li

M.Sc.
Institute for Communications Technology
Technische Universität Braunschweig
Schleinitzstraße 22 (Room 302)
38106 Braunschweig

zhengyang.li(at)tu-bs.de
phon.: +49 (0) 531 391 - 2443
fax: +49 (0) 531 391 - 8218

Publications: Google Scholar

ALL IN MULTI-MODAL LARGE LANGUAGE MODEL IN 2025! JOIN ME IN THIS JOURNEY!

Supervised Theses (Bachelor-, Master-, and Project-Thesis)

09. Fabian Bettels, Large World Model for Video Generation With Temporal Prediction of Image Tokens, 01/04/24-01/10/24, Joint supervision with Malte Stelzer

08. Pascal Reichert, An Efficient and Noise-Robust Audiovisual Encoder, 15/03/24-15/09/24

07. Shanhong Yin, Improving Noise Robustness of Audiovisual Speech Recognition by PixMix-Style Augmentation, 01/12/23-01/06/24

06. Moritz Möller, Few-Shot Training to Learn a Prompt Codebook for Contrastive Language-Image Pre-trained Models in Semantic Segmentation, 16/07/23-16/01/24, joint supervision with Björn Möller and Timo Bartels

05. Domenik Jaspers, Automatic Speech Recognition for German Children Speech:  Domain Generalization and Adaptation for Hybrid and End-to-End Approaches, 17/07/23-17/01/24

04. Patrick Blumenberg, Cross-Language Transfer Learning for Audiovisual Speech Recognition, 30/08/23-30/11/23, paper cooperated with Amazon AGI, to be presented at Interspeech2024

03. Thomas Graave, Cross-Language Transfer Learning for Audiovisual Speech Recognition, 01/02/23-01/08/23, paper cooperated with Amazon Alexa, presented at ASRU2023

02. Krishnan Jothi Ramalingam, A Comparison of Fully and Weakly Supervised Learning for Entity Recognition, 12/12/22-06/06/23, external master thesis supervised joinly with Mercedes-Benz

01. Chenwei Liang, Distill-AV-HuBERT: Audiovisual Representation Learning by Distillation Methods, 01/08/23-01/02/23, paper presented at Interspeech2023