Automatic Synchronisation Between Audio and Score Musical Description Layers

This work describes algorithms dedicated to score and audio alignment using the MX / IEEE P1599 format. The format allows description of the score, and management of synchronisation points, linking them with different versions of the performed music. An a

  • PDF / 488,551 Bytes
  • 11 Pages / 430 x 660 pts Page_size
  • 15 Downloads / 169 Views

DOWNLOAD

REPORT


bstract. This work describes algorithms dedicated to score and audio alignment using the MX / IEEE P1599 format. The format allows description of the score, and management of synchronisation points, linking them with different versions of the performed music. An algorithm is proposed here that allows alignment of an MX score and its execution, coded in PCM format, which produces an output for the MX Spine that contains synchronisation between notes and audio signal. The proposed architecture is based on two different steps: the first deals with the audio level and the extraction of features like pitch, onset and the like, while the second determines the alignment between the features extracted in the first step and the notes present in the MX score. For each step, different algorithms are proposed and discussed, and analysis and comparison of synchronisation capabilities are provided. Keywords: IEEE P1599, MX, synchronisation, PCM, MIDI.

1

Introduction

Contemporary digital music archives consist of huge collections of heterogeneous documents. For a music piece, an archive may contain corresponding scores in different versions, for example, voice and piano or orchestral, as well as several interpretations, e.g. played by different performers and recorded in diverse formats (CD recordings, MP3, FLAC and so on). The heterogeneity of music information makes retrieval hard to accomplish [1] and as a consequence many problems remain unsolved. One important problem that needs a solution is synchronisation, which requires the implementation of algorithms that automatically link different audio streams of the same piece to symbolic data formats representing the different scores. In particular, for the Notational and Audio layers in MX, synchronisation means that some algorithms, for a given event in some representation of the music score, are capable of determining the timing of the corresponding audio B. Falcidieno et al. (Eds.): SAMT 2007, LNCS 4816, pp. 200–210, 2007. c Springer-Verlag Berlin Heidelberg 2007 

Automatic Synchronisation Between Audio and Score Musical

201

events within an audio representation in some format. After execution of a synchronisation process in MX it is possible, for a given audio event, to find the correct position in the score, in the Notational and the Audio layers. As stated in [2]: ”‘Such synchronisation algorithms have applications in many different scenarios: following some score-based music retrieval, linking structures can be used to access some suitable audio CD accurately to listen to the desired part of the interpretation.”’This possibility represents a useful tool for music students, who can listen to music audio and, at the same time, see the corresponding notes. The MX spine and MX Music Events are the linking structure described in [2]. Furthermore, the MX linking structure can be useful for musicologists who can use synchronising algorithms to link the interpretation layer and the score layer, and then use MX for the investigation of agogics and tempo. In addition, temporal linking o