At present, Zih-Syuan Lin is a first-year PhD student at the Distributed Digital Music Archives & Libraries Lab, Schulich School of Music, McGill University. She was a research assistant at the Music and Culture Technology Lab, Institute of Information Science, Academia Sinica, Taipei, Taiwan, during 2023 to 2025. She graduated from National Tsing Hua University with a Master’s and a Bachelor’s degree in Industrial Engineering and Engineering Management.

During these six years of academic pursuit, she has taken various foundational music theory courses and completed the first Mandarin pop album as the arranger and producer of GuguDrive Band. As a composer, her instrumental works primarily consist of light music, aspiring to depict scenes through simple lines. As an arranger, she endeavors to create arrangements for a string quintet (or sextet) which could provide a pleasant experience for both the vocalist and accompaniment.

Her research interests accross music technology and machine learning, mainly focuses on Music Information Retrieval (MIR). She looks forwards to the real-world applications of various technologies as she experiences practical issues in the field of industrial engineering. For example, she and her teammates developed an algorithm for discriminating four-part writing based on tonal music theory. This aimed to alleviate the workload of teachers when grading assignments. Besides, she has experience working with a company that produces print circuit boards and developing a data-driven framework for improving product quality, utilizing regression models such as random forest and XGBoost.

She is also continuing previous research of texture classification, which aims to predict the role of each instrument at different time period in a symphony (whether it is a melody, accompaniment, or rhythm section) using piano roll information, by increasing the data volume and trying to improve the existing model accuracy. She hopes this research can aid in the structural analysis of music, such as phrase segmentation tasks. Additionally, she is interesting in music analysis, seeking to enhance the model's cognitive abilities in harmony rhythm and harmony progression, and she looks forward to considering multi-voice music recognition and generation in the future.