摘要
本論文主要探討基於ROS(Robot Operating System)的方式，自製一個智慧音樂伴奏系統。藉由先錄製一段吉他音樂，再將此聲音訊號做處理，分析出吉他的節奏(BPM, Beat Per Minute)，並把這些數據利用ROS的方式傳送到連接Arduino的馬達上，使得裝有馬達的鼓能夠配合吉他的節奏來演奏，像是一個無人樂團表演。 在音樂分析的方式主要是利用離散小波轉換的方式，讓我們可以同時分析聲音訊號在時域與頻域上的特徵值。而在資料的傳送上則是利用ROS的架構來與裝有Arduino的馬達做溝通，使其可以在同個網路下同時串聯許多樂器，而且可以根據使用者來加入自己喜歡的樂器，來達到即使只有一個人也可以完成音樂的練習及表演。此系統特色為鼓可以隨著吉他手的演奏速度快慢而改變打擊的節奏，有別於市面上只能以固定節奏打擊的機器人打鼓系統。同時，也設計了踏板讓吉他手可以依演奏需求做更換大鼓與小鼓的組合搭配，讓獨自一人也能擁有樂團的演出效果。
並列摘要
This project focuses on the creation of an intelligent music accompaniment system based on the ROS. By recording a piece of guitar music then processing the sound signal and analyzing the guitar rhythm (BPM, Beat Per Minute). By sending these data to the motor connected to the Arduino using ROS, the drums equipped with the motor can play with the rhythm of the guitar, like an unmanned band performance. The music analysis is mainly done by discrete wavelet conversion, which allows us to analyze the characteristic values of the sound signal in the time and frequency domain at the same time. In terms of data transmission, the ROS architecture is used to communicate with the motor so that many instruments can be connected to the same network at the same time, and the way it changes can be added according to the user's favorite instruments. Even if you are the only one, you can finish the music practice and performance.
參考文獻
延伸閱讀
- 簡瑜萱（2017）。基於ROS之人形機器人的影像定位與導航〔碩士論文，淡江大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0002-0708201714044600
- 陳建豪（2022）。使用人工智慧晶片實作之自動樂譜辨識與打擊樂演奏系統〔碩士論文，國立臺灣師範大學〕。華藝線上圖書館。https://doi.org/10.6345/NTNU202201582
- 鄭敬錡（2020）。基於ROS開發工業應用之無人搬運車安全及強健移動式機器人導航策略〔碩士論文，國立臺灣師範大學〕。華藝線上圖書館。https://doi.org/10.6345/NTNU202001147
- Ng, Y. M. (2006). Technological approach [master's thesis, The University of Hong Kong]. Airiti Library. https://www.airitilibrary.com/Article/Detail?DocID=U0029-1812201200013276
- Wang, L. C., Cao, Z. F., Cao, F., Qian, H. F., & Bao, H. Y. (2008). Biased Bit Commitment and Applications. Journal of Information Science and Engineering, 24(2), 441-452. https://doi.org/10.6688/JISE.2008.24.2.8