基于charRNN的复音音乐生成方法
|
摘要
在音乐生成过程中,charRNN方法只能对单音音乐进行训练,而不适用于多个乐器合奏的复音音乐。为使charRNN能适用于复音音乐,提出一种将MIDI音乐转换为一种基于一定语法规则的音乐描述语言的方法。利用charRNN完成文本训练,得到音乐生成模型,基于十二平均律方法获得音乐的统计特性,从而比较不同音乐片段间的差异。实验结果表明,该方法生成的音乐与真实音乐在结构和听感上比较相似,可用于多轨道复音音乐的自动生成。
In the music generation process,the charRNN method can only train monophonic music,and is not suitable for polyphonic music of multiple instrumental ensembles.In order to make charRNN suitable for polyphonic music,a method of converting MIDI music into a music description language based on certain grammatical rules is proposed.The text training is completed by using charRNN,thus obtaining a music generation model.The statistical properties of the music are obtained based on the theory of twelve-tone temperament method to compare the differences between the different pieces of music.Experimental results show that the music generated by this method is similar to the real music in structure and hearing,and can be used for automatic generation of multi-track polyphonic music.
In the music generation process,the charRNN method can only train monophonic music,and is not suitable for polyphonic music of multiple instrumental ensembles.In order to make charRNN suitable for polyphonic music,a method of converting MIDI music into a music description language based on certain grammatical rules is proposed.The text training is completed by using charRNN,thus obtaining a music generation model.The statistical properties of the music are obtained based on the theory of twelve-tone temperament method to compare the differences between the different pieces of music.Experimental results show that the music generated by this method is similar to the real music in structure and hearing,and can be used for automatic generation of multi-track polyphonic music.
引文
[1] 余立功,卜佳俊,陈纯.基于内外概率算法的音乐节奏自动生成[J].浙江大学学报(工学版),2005,39(12):1969-1972,1983.
[2] GRAVES A.Generating sequences with recurrent neural networks[EB/OL].[2018-02-01].https://arxiv.org/pdf/1308.0850.pdf.
[3] BAUM L E,PETRIE T,SOULES G,et al.A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains[J].The Annals of Mathematical Statistics,1970,41(1):164-171.
[4] HILLER L,ISAACSON L M.Experimental music composition with an electronic computer[M].Westport,USA:Greenwood Publishing Group Inc.,1979.
[5] ALLAN M,WILLIAMS C K I.Harmonising chorales by probabilistic inference[C]//Proceedings of the 17th International Conference on Neural Information Processing Systems.Cambridge,USA:MIT Press,2005:25-32.
[6] SIMON I,MORRIS D,BASU S.MySong:automatic accompaniment generation for vocal melodies[C]//Proceedings of the SIGCHI Conference on Human Factors in Computing Systems.New York,USA:ACM Press,2008:725-734.
[7] JACOB B L.Algorithmic composition as a model of creativity[J].Organised Sound,1996,1(3):157-165.
[8] BILOTTA E,PANTANO P,TALARICO V.Synthetic harmonies :an approach to musical semiosis by means of cellular automata[C]//Proceedings of the 7th International Conference on Artificial Life.Cambridge,USA:MIT Press,2002:153-159.
[9] 王存睿,段晓东,刘向东,等.基于Hilbert映射的元胞自动机音乐生成算法[J].微电子学与计算机,2010,27(1):5-8.
[10] DE LA PUENTE A O,ALFONSO R S,MORENO M A.Automatic composition of music by means of grammatical evolution[C]//Proceedings of Conference on APL.New York,USA:ACM Press,2002:148-155.
[11] LEWIS J.Algorithms for music composition by neural nets Improved CBR paradigms[EB/OL].[2018-02-04].https://quod.lib.umich.edu/cgi/p/pod/dod-idx/algorithms-for-music-composition.pdf?c=icmc;idno=bbp2372.1989.044;format=pdf.
[12] ECK D,SCHMIDHUBER J A first look at music composition using lstm recurrent neural networks[EB/OL].[2018-02-04].http://people.idsia.ch/～juergen/blues/IDSIA-07-02.pdf.
[13] LAMBERT A J,WEYDE T,ARMSTRONG N.Perceiving and predicting expressive rhythm with recurrent neural networks[EB/OL].[2018-02-01].http://openaccess.city.ac.uk/16489/.
[14] CHOI K,FAZEKAS G,SANDLER M.Text-based LSTM networks for automatic music composition[EB/OL].[2018-02-01].https://arxiv.org/pdf/1604.05358.pdf.
[15] 李雄飞,冯婷婷,骆实,等.基于递归神经网络的自动作曲算法[J].吉林大学学报(工学版),2018,48(3):866-873.
[16] HOCHREITER S,BENGIO Y,FRASCONI P,et al.Gradient flow in recurrent nets:the difficulty of learning long-term dependencies[EB/OL].[2018-02-01].https://www.bioinf.jku.at/publications/older/ch7.pdf.
[17] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[18] LI Rongjian,ZHANG Wenlu,SUK H I,et al.Deep learning based imaging data completion for improved brain disease diagnosis[C]//Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention.Berlin,Germany:Springer,2014:305-312.
[19] KLERK D D.Equal temperament[J].Acta Musicologica,1979,51(1):140-150.
[2] GRAVES A.Generating sequences with recurrent neural networks[EB/OL].[2018-02-01].https://arxiv.org/pdf/1308.0850.pdf.
[3] BAUM L E,PETRIE T,SOULES G,et al.A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains[J].The Annals of Mathematical Statistics,1970,41(1):164-171.
[4] HILLER L,ISAACSON L M.Experimental music composition with an electronic computer[M].Westport,USA:Greenwood Publishing Group Inc.,1979.
[5] ALLAN M,WILLIAMS C K I.Harmonising chorales by probabilistic inference[C]//Proceedings of the 17th International Conference on Neural Information Processing Systems.Cambridge,USA:MIT Press,2005:25-32.
[6] SIMON I,MORRIS D,BASU S.MySong:automatic accompaniment generation for vocal melodies[C]//Proceedings of the SIGCHI Conference on Human Factors in Computing Systems.New York,USA:ACM Press,2008:725-734.
[7] JACOB B L.Algorithmic composition as a model of creativity[J].Organised Sound,1996,1(3):157-165.
[8] BILOTTA E,PANTANO P,TALARICO V.Synthetic harmonies :an approach to musical semiosis by means of cellular automata[C]//Proceedings of the 7th International Conference on Artificial Life.Cambridge,USA:MIT Press,2002:153-159.
[9] 王存睿,段晓东,刘向东,等.基于Hilbert映射的元胞自动机音乐生成算法[J].微电子学与计算机,2010,27(1):5-8.
[10] DE LA PUENTE A O,ALFONSO R S,MORENO M A.Automatic composition of music by means of grammatical evolution[C]//Proceedings of Conference on APL.New York,USA:ACM Press,2002:148-155.
[11] LEWIS J.Algorithms for music composition by neural nets Improved CBR paradigms[EB/OL].[2018-02-04].https://quod.lib.umich.edu/cgi/p/pod/dod-idx/algorithms-for-music-composition.pdf?c=icmc;idno=bbp2372.1989.044;format=pdf.
[12] ECK D,SCHMIDHUBER J A first look at music composition using lstm recurrent neural networks[EB/OL].[2018-02-04].http://people.idsia.ch/～juergen/blues/IDSIA-07-02.pdf.
[13] LAMBERT A J,WEYDE T,ARMSTRONG N.Perceiving and predicting expressive rhythm with recurrent neural networks[EB/OL].[2018-02-01].http://openaccess.city.ac.uk/16489/.
[14] CHOI K,FAZEKAS G,SANDLER M.Text-based LSTM networks for automatic music composition[EB/OL].[2018-02-01].https://arxiv.org/pdf/1604.05358.pdf.
[15] 李雄飞,冯婷婷,骆实,等.基于递归神经网络的自动作曲算法[J].吉林大学学报(工学版),2018,48(3):866-873.
[16] HOCHREITER S,BENGIO Y,FRASCONI P,et al.Gradient flow in recurrent nets:the difficulty of learning long-term dependencies[EB/OL].[2018-02-01].https://www.bioinf.jku.at/publications/older/ch7.pdf.
[17] HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[18] LI Rongjian,ZHANG Wenlu,SUK H I,et al.Deep learning based imaging data completion for improved brain disease diagnosis[C]//Proceedings of International Conference on Medical Image Computing and Computer-Assisted Intervention.Berlin,Germany:Springer,2014:305-312.
[19] KLERK D D.Equal temperament[J].Acta Musicologica,1979,51(1):140-150.