- 期刊
應用調變頻譜分析於音樂樂器之自動辨識
Automatic Musical Instrument Recognition Using Modulation Spectral Analysis
摘要
本論文應用調變頻譜分析於音樂樂器自動辨識，首先我們自樂器音符中擷取頻帶能量調變頻譜特徵、頻帶平均能量調變頻譜特徵、頻帶能量標準差調變頻譜特徵、擴充頻帶能量調變頻譜特徵、擴充頻帶平均能量調變頻譜特徵和擴充頻帶能量標準差調變頻譜特徵作為音符之特徵，然後將測詴音符與訓練音符用歐基里德距離來計算兩個音符之間的距離，我們取音符距離最小者的樂器類別作為辨識之類別。我們實驗的樂器種類為8種包含：中音薩克斯風、低音提琴、大提琴、長笛、雙簧管、小號、中提琴和小提琴，經由對這8種樂器進行自動辨識之實驗後，其中發現擴充頻帶能量標準差調變頻譜特徵的辨識率最好，對於IOWA及RWC資料庫其辨識率分別為92.72%及92.51%。
並列摘要
In this paper, modulation spectral analysis is employed to extract discriminated features for musical instruments recognition. First we apply modulation spectral analysis on subband energy, to extract variant modulation spectral features, including subband average energy, subband energy standard deviation, extended subband energy, extended subband average energy, and extended subband energy standard deviation. Finally, the Eucidean distance is used to evaluate the distance between test note and each train note. In our experiments, eight instrumental classes, including alto saxphone, bass, cello, flute, oboe, trumpet, violin, and viola, were used to evaluate the performance. The proposed modulation spectral analysis of extended subband energy standard deviation feature achieve the highest recognition accuracy of 92.72% and 92.51% for IOWA and RWC databases.
延伸閱讀
- Lai, C. C. (2011). 調變頻譜正規化法於強健語音辨識之研究 [master's thesis, National Chi Nan University]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0020-1207201117352400
- 葉子豪（2013）。調變頻譜冪次展開法於強健語音辨識之研究〔碩士論文，國立暨南國際大學〕。華藝線上圖書館。https://doi.org/10.6837/NCNU.2013.00097
- Sun, L. C. (2011). 基於調變頻譜等化法之強健性語音辨識技術 [doctoral dissertation, National Taiwan University]. Airiti Library. https://doi.org/10.6342/NTU.2011.00539
- 朱紋儀（2011）。調變頻譜特徵正規化於強健語音辨識 之研究〔碩士論文，國立臺灣師範大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0021-1610201315254737
- 王炫凱（2009）。Estimation of User’s Affective Response on MusicContents Using Real-Time Analysis System of Physiological Signals〔碩士論文，國立臺灣大學〕。華藝線上圖書館。https://doi.org/10.6342/NTU.2009.01669