發(fā)布時(shí)間：2018-01-10 04:25

  本文關(guān)鍵詞：基于智能可穿戴設(shè)備的樂音對(duì)比算法　出處：《應(yīng)用科學(xué)學(xué)報(bào)》2017年06期 　論文類型：期刊論文

  更多相關(guān)文章： 樂音對(duì)比 次諧波求和 基頻提取 實(shí)時(shí)性 可穿戴設(shè)備

【摘要】：提出一種基于智能可穿戴設(shè)備的樂音對(duì)比算法.該算法根據(jù)樂音信號(hào)特性,用平均幅值法定位音符起始點(diǎn),由短時(shí)傅里葉變換獲取頻域信息;將原始頻譜用次諧波求和法與最小方差法提取基頻信息,進(jìn)而精確定位樂音位置;通過計(jì)算演奏樂音與模板樂音時(shí)頻域特征的歐氏距離,實(shí)現(xiàn)實(shí)時(shí)的樂音對(duì)比.該方法計(jì)算復(fù)雜度極低,實(shí)時(shí)性很好,準(zhǔn)確率高.仿真實(shí)驗(yàn)表明,該方法適用于智能可穿戴設(shè)備的樂音對(duì)比.
[Abstract]:An intelligent wearable device - based tone contrast algorithm is proposed . Based on the characteristics of musical tone signal , the initial point of note is determined by means of mean amplitude method . The frequency domain information is obtained by short - time Fourier transform . The original frequency spectrum is extracted by the sub - harmonic summation method and the minimum variance method to accurately locate the tone position . The method is very low in calculation complexity , good in real - time performance and high in accuracy . The simulation experiment shows that the method is suitable for the tone contrast of intelligent wearable devices .

【作者單位】： 上海大學(xué)通信與信息工程學(xué)院;
【基金】：國(guó)家自然科學(xué)基金面上項(xiàng)目(No.61571282)資助
【分類號(hào)】：TN912.3;TP368.33
【正文快照】： 近年來,音樂識(shí)別技術(shù)發(fā)展迅速,在眾多領(lǐng)域中應(yīng)用廣泛[1].同時(shí),可穿戴設(shè)備已經(jīng)完全融入人們的生活,并以自身智能、便捷、高效等特點(diǎn)成為許多人生活中重要的組成部分[2],在運(yùn)動(dòng)健康、智能控制、生活輔助等方面有很好的發(fā)展,這就促進(jìn)了有音樂陪練功能的智能可穿戴設(shè)備的誕生.但問，

本文編號(hào)：1403865