鄭士康臺灣大學:電機工程學研究所林睿敏Lin, Ruei-MinRuei-MinLin2007-11-262018-07-062007-11-262018-07-062006http://ntur.lib.ntu.edu.tw//handle/246246/53374本論文提出了一個新的音高辨識方法:利用簡單的耳蝸物理模型,可以讓電腦模擬人耳聽覺的產生,來辨識音高。傳統音高辨識方法主要分為兩大類:第一類型方法利用簡單的自相關函式,將一段聲音波形輸入函式後,經由簡單的運算,可以迅速得到主要頻率音高;第二類型方法則是將一段時域資料的聲音波形,經過傅立葉轉換後,轉成頻域資料得到頻譜,再分析頻譜得到音高。本論文藉由一個簡單的耳蝸物理模型,可以直接利用時域資料去振動耳蝸中的基底膜,再藉由分析基底膜的振動情形抓出音高。不同於第一類方法只能抓到一個主要的頻率,我們的方法,因為整條基底膜的彈性並不一致,所以可以同時抓出各個頻率的組成大小。另外,少了頻譜轉換的步驟,因此我們的方法,運算速度比起第二類方法快速許多。In this paper, an algorithm for pitch recognition is designed. This algorithm is based on a simplified cochlear model. The traditional methods are mainly divided into two categories: one is to utilize and analyze the amplitude of sound in time domain directly; the other is to transform the sound into the frequency domain first, and then do some analysis to recognize the pitch. The operation amount in time domain is relatively small, but mostly it can only detect a single frequency. The second type of methods needs to do the transform first, so the speed is relatively slow. After getting the frequency spectrum, we can apply some algorithm to do the pitch recognition. My algorithm, which is called CM (Cochlear Model), combines the advantages of above-mentioned two kinds of methods. CM utilizes the amplitude of sound directly. Through the simple cochlea physical model, the vibration situation of the BM(basement membrane) in the cochlea can tell the pitch. For the elasticity in the BM is not uniform, we can tell more than one single frequency at the same time.目錄 摘要……………………………………………………………………………………i Abstract………………………………………………………………………………..ii 目錄…………………………………………………………………………………...iii 表目錄………………………………………………………………………………....v 圖目錄………………………………………………………………………………...vi 第一章 緒論…………………………………………………………………………..1 1.1 本章概要……………………………………………………………….1 1.2 問題描述 ………………………………………………………………1 1.3 研究動機與目的……………………………………………………….2 1.4 音高辨識傳統方法簡介……………………………………………….2 1.5 利用耳蝸物理模型為基礎的音高辨識方法簡介…………………….3 1.6 論文架構……………………………………………………………….3 第二章 傳統音高辨識方法介紹……………………………………………………..4 2.1 本章概要 ………………………………………………………………4 2.2 時域音高辨識方法…………………………………………………….4 2.2.1 (Zero Crossings)………………………………………....4 2.2.2 (Autocorrelation Function)……………………………6 2.2.3 (Average Magnitude Difference Function) …………7 2.2.4其他時域音高辨識方法………………………………………...8 2.2.5 時域辨識音高方法小結………………………………………..9 2.3 頻域音高辨識方法 ……………………………………………………9 2.3.1 Hamming框化…………………………………………………..9 2.3.2 (Discrete Fourier Transform)…………………………10 2.3.3 (Constant Q Transform)..................................................15 2.3.4 其他頻域音高辨識方法............................................................17 2.3.5 頻域音高辨識方法小結............................................................18 2.4 其他音高辨識方法 ..............................................................................18 2.5 總結.......................................................................................................18 第三章 聽覺的產生…………………………………………………………………20 3.1 本章概要……………………………………………………………...20 3.2 外耳 (The Outer Ear)…………………………………………….......20 3.3 中耳 (The Middle Ear) ………………………………………………20 3.4 內耳 (The Inner Ear) ………………………………………………...21 3.5 總結………………………………………………....………………...21 第四章 耳蝸物理模型為基礎的音高辨識方法…………………………………..22 4.1 本章概要.……………………………………………………………..22 4.2 簡化的耳蝸物理模型..….……………………………………………22 4.3 相關數學公式推導 .….………………………………………………23 4.4 合理的參數設定.….………………………………………………….25 4.5 程式實作 .….…………………………………………………………27 4.6 總結 .….………………………………………………………………29 第五章 結果討論……………………………………………………………………30 5.1 本章概要…………………………………………………….….….…30 5.2 單音音高辨識結果比較 ……………………………………………..30 5.2.1 正弦波辨識結果比較…………………………………………30 5.2.2 方波辨識結果比較……………………………………………36 5.2.3 吉他單音辨識結果比較………………………………………41 5.2.4 小結……………………………………………………………47 5.3 複音(吉他C和弦)辨識結果比較………………………….….….48 5.4 與 的比較……………………………………………………50 5.5 總結 …………………………………………………………………..52 第六章 結論…………………………………………………………………………53 參考文獻……………………………………………………………………………..541215661 bytesapplication/pdfen-US音高辨識耳蝸耳蝸模型基底膜pitchpitch trackingpitch recognitionpitch determination algorithmPDAcochleacochlear modelbasement membraneBM耳蝸物理模型為基礎的音高辨識方法Pitch Recognition Based on Cochlear Modelthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/53374/1/ntu-95-R93921026-1.pdf