Preparing for TreeTorika : Computer-Assisted Analysis of Mao’s oratory

This paper examines computer-assisted analysis techniques in extracting musical features from a recording of a speech by Mao Zedong. The data was used to prepare compositional material such as global form, melody, harmony and rhythm for TreeTorika for chamber orchestra. The text focuses on large-scale segmentation, melody transcription, quantification and quantization. It touches upon orchestration techniques but does not go into other aspects of the work such as constrained-based rhythm development. Automatic transcription of the voice was discarded in favour of an aurally based method supported by tools in Amadeus and Max/MSP. The data were processed in OpenMusic to optimise the notation with regards to accuracy and readability for the musicians. The harmonic context was derived from AudioSculpt partial tracking and chord-sequence analyses. Finally, attention will be given to artistic and political considerations when using recordings of such an intensively disputed public figure as Mao.