1. 課程說明(course desc<x>ription)：
<br>To introduce application of signal processing methods in music, language, and 
<br>other sounds in daily life, including two major parts: sound analysis and sound 
<br>synthesis. After taking this course, students should be familiar with the core 
<br>techniques and background knowledge in this field. Also, current research topics 
<br>will be selectively covered. Final projects are required and students are 
<br>encouraged to be creative. Prerequisite: Signals and Systems, Linear Algebra, or 
<br>consent by the instructor. Undergraduates in junior or senior years are also 
<br>welcome to take this course.
<br>
<br>AI policies:
<br>有條件開放，請註明如何使用生成式AI於課程產出
<br>
<br>生成式人工智慧倫理聲明 「有條件開放，請註明如何使用生成式AI於課程產出」
<br>
<br>基於透明與負責任的原則，本課程鼓勵學生利用AI進行協作或互學，以提升本門課產出品質。根據本校公布之
<br>「大學教育場域AI協作、共學與素養培養指引」，本門課程採取有條件開放，以下說明如何使用生成式AI於課
<br>程產出
<br>
<br>學生須於課堂作業或報告中的「標題頁註腳」或「引用文獻後」簡要說明如何使用生成式AI進行議題發想、文句
<br>潤飾或結構參考等使用方式。若經查核使用卻無在作業或報告中標明，教師、學校或相關單位有權重新針對作業
<br>或報告重新評分或不予計分。
<br>本門課授課教材或學習資料若有引用自生成式AI，教師也將在投影片或口頭標注。
<br>修讀本課程之學生於選課時視為同意以上倫理聲明。
<br>
<br>2.  Textbook: None.
<br>
<br>3. References:
<br>I plan to focus on speech and human voice. Below is a good reference I've found,
<br>
<br>L. R. Rabiner and R. W. Schafer, "Introduction to digital speech processing,"
<br>Foundations and Trends in Signal Processing, Vol. 1, Nos. 1–2 (2007) 1–194. 
<br>
<br>You may find its PDF online.
<br>https://www.nowpublishers.com/article/Details/SIG-001
<br>
<br>Otherwise, lecture notes and papers for group discussion will be handed out 
<br>throughout the semester. 
<br>
<br>4. Teaching methods: a mixture of lectures and student-reporting. We embrace a 
<br>problem-ba<x>sed learning paradigm. Followings are the course ob<x>jectives -- we want 
<br>students to cultivate these abilities: 
<br>(1) to write an algorithm and implement signal processing methods from their 
<br>mathematical desc<x>riptions (a set of equations). 
<br>(2) to analyse human voice signals, modify parametric representations, and 
<br>resynthesize to create special effects.
<br>(3) to relate voice processing methods to underlying assumptions about human voice 
<br>perception and production  
<br>(4) to choose appropriate performance evaluation methods for quality of 
<br>synthesized voice, including subjective and ob<x>jective ones.
<br>
<br>
<br>5. Syllabus (Tentative):
<br>For your reference, in Spring 2021, we covered the following topics
<br>-- Basic Fourier transform and digital filtering
<br>-- Linear prediction voice processing
<br>-- sinusoidal modeling-ba<x>sed voice processing
<br>-- sine + noise decomposition and audio data compression
<br>-- vocoders
<br>
<br>6. Evaluation (Tentative)
<br>-- 8 homework (64%): computer-ba<x>sed.
<br>-- Final Group Project (20%)
<br>-- 4 Reading digest reports (8%)
<br>-- Class-participation (8%)
<br>
<br>
<br>7. Webpage: 
<br>Course materials will be distributed through NTHU eeclass system.
<br>
<br>Please activate your account at
<br>https://eeclass.nthu.edu.tw/
<br>if you have not done so. 
<br>