Abstract：A chord sequence recognition algorithm based on Liquid State Machine (LSM) is presented. Firstly,the music signal is segmented and Pitch Class Profile feature is extracted for every frame. Then,a LSM model is achieved after training. Two kinds of Bizarre Chord,chord appears probability vector and chord transformation matrix,are presented to post-process the chord sequence outputted by LSM. 8 sets of experimental data from neural network model,hidden Markov mode,echo state network model and LSM model show that the LSM gets a good performance,and the post-processing method also effectively improves the recognition accuracy.
[1] Fujishima T. Realtime Chord Recognition of Musical Sound: A System Using Common Lisp Music // Proc of the International Computer Music Conference.Beijing,China,1999: 464-467 [2] Harte C A,Sandler M B. Automatic Chord Identification Using a Quantized Chromagram // Proc of the 119th Audio Engineering Society Convention.New York,USA,2005: 6412-6419 [3] Ellis D P W,Poliner G E. Identifying Cover Songs with Chroma Features and Dynamic Programming Beat Tracking // Proc of the IEEE International Conference on Acoustics,Speech and Signal Processing. Hawaii,USA,2007: 1429-1432 [4] Ueda Y,Uchiyama Y,Nishimoto T,et al. HMM-Based Approach for Automatic Chord Detection Using Refined Acoustic Features // Proc of the IEEE International Conference on Acoustics,Speech and Signal Processing. Dallas,USA,2010: 5518-5521 [5] Harte C,Sandler M,Gasser M. Detecting Harmonic Change in Musical Audio // Proc of the 1st ACM Workshop on Audio and Music Computing Multimedia. Santa Barbara,USA,2006: 21-26 [6] Uchiyama Y,Miyamoto K,Nishimoto T,et al. Automatic Chord Detection Using Harmonic Sound Emphasized Chroma from Musical Acoustic Signal // Proc of the International Conference of Music Information Retrieval. Philadelphia,USA,2008: 901-902 [7] Lee K. A System for Acoustic Chord Transcription and Key Extraction from Audio Using Hidden Markov Models Trained on Synthesized Audio. Ph.D Dissertation. Stanford,USA: University,2008 [8] Lee K,Slaney M. Acoustic Chord Transcription and Key Extraction from Audio Using Key-Dependent HMMs Trained on Synthesized Audio. IEEE Trans on Audio,Speech,and Language Processing,2008,16(2): 291-301 [9] Sun Jiayin,Li Haifeng,Lei Li. Music Chord Real-Time Perception Based on the Artificial Neural Network // Proc of the National Conference on Man-Machine Speech Communication. Urumqi,China,2009: 11-16 [10] Gerhard D,Zhang Xinglin. Chord Analysis Using Ensemble Constraints // R s Z W,Wieczorkowska A A,Alicja W,eds. Advances in Music Information Retrieval. Berlin: Springer-Verlag,2010: 119-142 [11] Maass W,Natschlger T,Markram H. Real-Time Computing without Stable States: A New Framework for Neural Computation Based on Perturbations. Neural Computation,2002,14(11): 2531-2560 [12] Verstraeten D,Schrauwen B,Stroobandt D. Isolated Word Recognition Using a Liquid State Machine // Proc of the 13th European Symposium on Artificial Neural Networks. Brugge,Belgium,2005: 435-440 [13] Jaeger H. The Echo State” Approach to Analyzing and Training Recurrent Neural Networks with an Erratum Note. Technical Report,GMD148. Bonn,Germany: German National Research Center for Information Technology,2001 [14] Verstraeten D,Schrauwen B,d′ Haene M,et al. An Experimental Unification of Reservoir Computing Methods. Neural Networks,2007,20(3): 391-403 [15] Burgoyne J A,Wild J,Fujinaga I. An Expert Ground-Truth Set for Audio Chord Recognition and Music Analysis // Proc of the 12th International Society for Music Information Retrieval Conference. Miami,USA,2011: 633-638
2013, Vol. 26 
