Project # 1.1 : audio skimming

Within the numediart/HyForge research program, this project aims at studying techniques and developing software prototypes for interactive skimming audio contents. After surveying the stateof- the-art techniques for time-scale modifications of audio signals, we adopt a phase vocoding method. Some limitations of this method are identified and solutions are proposed. More especially, we derive an algorithm that allows adapting continuously the skimming parameters (i.e., the position in the sound excerpt and the playing speed) and investigate the use of phase locking and transient detection in order to compensate for spectral smearing and phasiness. Besides, we develop a software prototype as external objects within the PureData environment, namely a signal processing object and a visualization object that computes the audio output and visualizes the signal waveform, respectively, while skimming an audio file. This prototype performs efficiently in realtime and can be steered with various controllers through OSC support. It definitively provides a suitable platform for testing future developments.