基於樂理運用音高週期偵測演算法在和弦辨識之研究

本論文中，提出一種運用音高週期偵測的演算法，以音樂理論為基礎在聲音訊號中偵測音高並做和弦辨識的研究，系針對不同的音高週期偵測演算法，加以深入探討，找尋出最適合的方法並提出改善。在這篇論文中，我們提出一種音高偵測的方式，針對音樂的單音和多音的和弦去做辨識。音高週期偵測可以在頻域上(Frequency Domain)與時域上(Time Domain)做音訊偵測，當我們辨識單音時，我們使用AMDF的方法在時域上做偵測。而在辨識多音時，則提出一個值觀的音高偵測方法，在頻域上做偵測。首先，我們先建立音樂檔案，並存成wave檔。第二步，我們做音樂的音高特性擷取。聲音信號通過快速傅立葉轉換(Fast Fourier Fransform)後，會將信號從時域轉換到頻域，捉取基頻上的峰音去判斷識別音高。我們將以音樂理論為基礎對和弦做分析，不同的和弦會影響音樂的結構，我們希望能利用和弦富含音樂性的特性，來代表音樂的主要內涵。在這篇論文中，我們提出以下的貢獻： 1. 本論文藉由聲音訊號的處理，有效的辨識音高，有助於學音樂的人學習效率提升。 2. 讓對於學習鋼琴、吉他、以及學習各種樂器的人，利用辨識出的和弦結果，做為旋律上的伴奏，在學習上不單調乏味。 3. 本論文所提的音高偵測演算法結合樂理，了解音樂感知這方面對樂曲的影響，提升辨識的正確性。

關鍵字

音高偵測；和弦辨識

並列摘要

This thesis presents an algorithm used in pitch period detection to detect the pitch in the audio signals and chords recognition based on musical theory. We propose the way of a pitch detection that focuses on recognition for music monophonic and polyphonic chords. The pitch period detection contains the frequency domain audio detection and time domain audio detection. When we identify the monophonic, it uses the AMDF (Average Magnitude Difference Function) in the time domain detection. When we identify the polyphone, we propose a pitch detection method in the frequency domain detection. First, We have to establish the music files. We use music conversion software to transfer music to wav files. Second, we will carry out the capture of musical pitch characteristics. In the polyphonic detection section, we use the fast fourier transform to process the audio signal convert it from the time domain to the frequency domain. Through this way, we will obtain the fundamental frequency and catch the peak for a pitch identification. We recognize the polyphonic chord analysis based on music theory. The different chord will affect the architecture of the music, so we hope to use the characteristics of the chords rich in musical to representative the main connotation of music. Finally, we present our simulation and results. In this thesis, the following contributions are gained: (1) In this thesis, by the audio signal processing will identify the pitch and melody effectively, help the musical study to learn it efficiency. (2) Use these results of the chord identification as the accompaniment of the melody let the people to learning piano, guitar, and learn a variety of musical instruments, and not monotonous on learning. (3) The pitch detection algorithms combine with music theory mentioned in this thesis, not only enhance the correctness of the identification but also help people to understand the music perception in this area of music.

並列關鍵字

Pitch Detection ； Chord Recognition

參考文獻

[1] Goto, M., “A Real-time Music Scene Description System: Predominant-F0 Estimation for Detecting Melody and Bass Lines in Real-world Audio Signals,” SpeechCommunication (ISCA Journal), Vol.43, No.4, pp.311-329, September 2004.

[3] Goto, M., “A Predominant-F0 Estimation Method for CD Recordings: MAP Estimation using EM Algorithm for Adaptive Tone Models,” Proceedings of the 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2001), pp.V-3365-3368, May 2001.

[4] Goto, M., “A Robust Predominant-F0 Estimation Method for Real-time Detection of Melody and Bass Lines in CD Recordings,” Proceedings of the 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2000), pp.II-757-760, June 2000.

[7] Scheirer, E. D., “Tempo and beat analysis of acoustic musical signals,”J. Acoust. Soc. Amer., vol. 103, no. 1, pp. 588–601, 1998.

[9] T. Marta, T. Raquel, K.Thomas, “Mood-based navigation through large collections of musical data”, 2004

國際替代計量

基於樂理運用音高週期偵測演算法在和弦辨識之研究

全文下載

主題瀏覽