课 程 内 容 【The Schedule of Each Day】 Monday Introduction and background Marrian information processing framework & Acoustics & Signal processing Tuesday Physiological and psychoacoustic basis of audition Physiological basis & Psychoacoustics & Auditory scene analysis & Real-world audition Wednesday Fundamentals of computational audition Auditory representations & MultiPitch tracking & Auditory segmentation & Ideal binary mask as CASA goal Thursday Computational auditory scene analysis Sound separation problem & Traditional CASA methods & Classification-based methods & Binaural segregation Friday Pattern recognition and learning Automatic speech recognition & Learning methods: MLP, SVM, and DNN & Overall discussion: Neural mechanisms of audition 1:30~3:30 p.m everyday Discussion Time 【课程内容简介】 【中文】这个为期一周的短期课程将会涵盖auditory perception and processing的计算模型和算法。我们将会涉及的话题包括：听觉场景分析, 声高分析, 声音定位, 语音和音乐处理 (增强, 隔离, 和录音), 语音和发言者识别, 和声学建模。这门课的的目标在于让学生理解一些声学在生理学或者心理声学上的现象，学习一些声学或听觉分析方面上的现代理论，模型和算法，并最终能解决听觉和语音处理当中的实际问题。 【En】This one-week, short-term course covers computational models and algorithms for auditory perception and processing. Topics include auditory scene analysis, pitch analysis, sound localization, speech and music processing (enhancement, segregation, and transcription), speech and speaker recognition, and modeling of the auditory system. The objectives of this course include understanding a set of physiological/psychoacoustic phenomena of audition, comprehension of modern theories, models, and algorithms for acoustic/auditory analysis, and being able to solve practical problems in audio and speech processing.