CV
Wei Zeng's CV (Updated 2026 Jan)
Basics
| Name | Wei Zeng |
| Label | Ph.D. Candidate, Sound and Music Computing |
| w.zeng@u.nus.edu | |
| Phone | +65 9379-9504 |
| Url | https://wei-zeng98.github.io/ |
| Summary | Ph.D. candidate at NUS SMC Lab focusing on automatic music transcription, expressive performance rendering, and singing voice synthesis. |
Education
-
2023.08 - 2027.05 Ph.D.
National University of Singapore
Computer Science
-
2021.08 - 2023.06 M.Eng.
National University of Singapore
Mechanical Engineering
-
2016.09 - 2020.06 B.Eng. / B.A.
Shanghai Jiao Tong University
Energy & Power Engineering (Major); Musicology (Minor)
Publications
-
2026 Bridging Piano Transcription and Rendering via Disentangled Score Content and Style
ICLR
Wei Zeng, Junchuan Zhao, Ye WangUnified transformer Seq2Seq jointly models APT and EPR by disentangling score content and performance style; diffusion-based style recommender enables alignment-free training and style transfer.
-
2025 Peransformer: Improving Low-informed Expressive Performance Rendering with Score-aware Discriminator
APSIPA ASC
Xian He, Wei Zeng, Ye WangScore-aware discriminator and transformer framework enhance expressive performance rendering when score information is sparse.
-
2024 Automatic Lyric Transcription and Automatic Music Transcription from Multimodal Singing
ACM TOMM
Xiangming Gu, Longshen Ou, Wei Zeng, Jianan Zhang, Nicholas Wong, Ye WangMultimodal ALT/AMT with audio and lip video using residual cross-attention; first N20EM dataset and SSL encoders yield noise-robust lyric and note transcription.
-
2024 End-to-End Real-World Polyphonic Piano Audio-to-Score Transcription with Hierarchical Decoding
IJCAI
Wei Zeng, Xian He, Ye WangHierarchical decoder predicts bar-level key/time and note content; two-stage training with EPR pretraining then human-performance finetuning bridges synthetic–real gap.
-
2023 Elucidate Gender Fairness in Singing Voice Transcription
ACM MM
Xiangming Gu, Wei Zeng, Ye WangReveals gender bias in singing voice transcription from pitch distributions; adversarial attribute predictor with note-conditioned alignment cuts bias >50% with minimal accuracy loss.
Research
-
2025.01 - 2025.09 SMC Lab, NUS
Zero-Shot Singing Voice Synthesis with Melody Control
Proposed CoMelSinger (MaskGCT-based zero-shot SVS) with disentangled melody control and contrastive learning; added encoder-only SVT for pitch supervision.
-
2024.09 - 2025.05 SMC Lab, NUS
Joint Piano Transcription and Performance Rendering
Unified transformer Seq2Seq for APT+EPR; alignment-free formulation and diffusion style recommendation.
-
2023.08 - 2024.02 SMC Lab, NUS
Hierarchical Audio-to-Score Transcription for Piano
End-to-end Seq2Seq with hierarchical decoder and two-stage training using EPR pretraining then real-perf finetuning.
Projects
- 2020.11 - 2021.04
FAISS Semantic Paper Retrieval
Shanghai Artificial Intelligence Research Institute, Shanghai
Built MySQL schema, ingested ~150M paper records (MAG/AMiner), fine-tuned BERT on abstracts/categories, and deployed FAISS-based semantic search.
Skills
| Programming | |
| Python | |
| MATLAB | |
| C++ |
Languages
| Mandarin | |
| First language |
| English | |
| TOEFL 102; GRE 324 |
Miscellaneous
| Service | |
| Student Area Search Committee (SoC, AY2025) | |
| Reviewer: TASLP, ICLR, etc. |
| Teaching | |
| IT5005 Artificial Intelligence (2024-2026) | |
| CS5647 Sound and Music Computing (2025) |
Awards
- 2025.06.01
- 2025.05.01
Teaching Fellowship of SoC
- 2024.06.01
Research Achievement Award of SoC
- 2020.06.01
Academic Excellence Scholarship of SJTU