User:Yunwang

Yun Wang (Maigo)

Yun Wang is my real name and Maigo is my nickname. You can call me either Yun or Maigo.

Who I am and why I'm here

I graduated from the Department of Electronic Engineering of Tsinghua University in 2010. Now I am a second-year master student at LTI, CMU.

I am interested in technologies about speech, natural language, and music. Currently I'm working with Prof. Qin Jin on speaker identification.

I'm taking this course because I've taken Language & Statistics, which gave me inspiring insights into language.

Publications

• Yun Wang and Zhijian Ou, "Combining HMM-based melody extraction and NMF-based soft masking for separating voice and accompaniment from monaural audio", Proceedings of ICASSP, pp. 1-4, May 2011.
• Angeliki Metallinou, Athanassios Katsamanis, Yun Wang and Shrikanth Narayanan. "Tracking changes in continuous emotion states using body language and prosodic cues", Proceedings of ICASSP, pp. 2288-2291, May 2011.

Superpowers

• I know 6 languages: Chinese, English, Japanese, Korean, Spanish, and French.

Random notes

• "Maigo" is a Japanese word meaning "lost child".

• You can recognize me by the red Mario hat.

Course Project

I'm working with Wangshu Pang and Matt Gardner on Improving Knowledge-Based Weakly Supervised Information Extraction.

Writeup Assignments

Papers

• (September) Grenager et al, ACL 2005: Unsupervised Learning of Field Segmentation Models for Information Extraction
• (September) Haghighi and Klein, ACL 2006: Prototype-Driven Learning for Sequence Models
• (September) Berg-Kirkpatrick et al, ACL 2010: Painless Unsupervised Learning with Features (On unsupervised learning with featurized HMMs)
• (October) McDonald et al, ACL 2005: Non-Projective Dependency Parsing Using Spanning Tree Algorithms
• (November) Gunawardana et al, ICSCT 2005: Hidden Conditional Random Fields for Phone Classification
• (November) Chelba and Acero, EMNLP 2004: Adaptation of Maximum Entropy Capitalizer: Little Data Can Help A Lot

Methods

• (September) Featurized HMM
• (October) Stacked Sequential Learning