【徵才单位】中研院资创中心 音乐与音讯运算实验室 Music & Audio Computing Lab, Research Center for IT Innovation, Academia Sinica http://mac.citi.sinica.edu.tw/ 【职务名称】博士後 1名 【研究内容】数位信号处理、机器学习、自然语言处理、社群多媒体相关研究 【工作待遇】依中研院/科技部规定 NTD 56,650 起 可议 享劳健保与年终奖金、年终1.5个月 【聘 期】1. 随到随审，可随时起聘 2. 一年一聘 3. 亦接受研发替代役申请 【工作地点】台北市南港区 中研院资创中心 http://www.citi.sinica.edu.tw/ 【应徵条件】国内外 电机/资讯 博士学历 【需求研究专长】 1. natural language processing 2. social multimedia 3. signal processing 4. machine learning 5. information retrieval 【具体研究内容】 参与以下两研究计画其中之一 1. Cross-cultural Analysis of Music Perception for Culture-aware Recommendation The global and extensive use of social media bears an unprecedented amount of personal and cultural information, which can, to some extent, be uncovered by state-of-the-art methods from machine learning and social data science. In combination with the likewise omnipresent consumption and enjoyment of music around the world, it has become possible to research in a multifaceted and large scale manner cultural similarities and differences of music listening behavior. In this project, we set forth to analyze millions of music-related microblogs collected from Twitter, to investigate how listening habits, music preferences, and music perception differ across cultures, and how to model such cultural differences over time, and how to exploit these differences to improve the state-of-the-art in music recommendation. This is a joint project with universities in Taiwan, Hong Kong, and Austria. 2. Complex-Valued Signal Processing and Feature Learning For Music Information Retrieval Content analysis of polyphonic music is arguably one of the most challenging tasks in computer audition, as it tackles the complexity of sound mixtures with overlapping harmonics components spreading over a wide frequency range. This project focuses on two research directions to overcome these challenges. The first is to find novel signal representations for music by using advanced time-frequency analysis and complex-valued signal processing techniques. The second is to design robust feature learning methods based on deep learning and sparse coding techniques. The applications we are interested in including automatic music transcription, source separation, performance analysis, score following, and music education. 【联络方式】杨奕轩副研究员 http://www.citi.sinica.edu.tw/pages/yang/ [email protected] 来信请请附履历、动机、代表着作 【相关连结】 https://tmacw16.wordpress.com/ http://dbis-nowplaying.uibk.ac.at/ http://c4dm.eecs.qmul.ac.uk/ismir15-amt-tutorial/ https://ccrma.stanford.edu/workshops/music-information-retrieval-mir-2015 --

※ 发信站: 批踢踢实业坊(ptt.cc), 来自: 60.199.29.165 ※ 文章网址: https://webptt.com/cn.aspx?n=bbs/AfterPhD/M.1455330335.A.EB6.html