AUDIO MELODY EXTRACTION FOR MIREX 2008

We submit a method for the MIREX 2008 task “Audio Melody Extraction”. The method first transcribes input audio file into MIDI notes and then estimates fundamental frequency (F0) trajectory for each note. The melody transcription method uses a frame-wise pitch salience estimator to measure the strength of different fundamental frequencies. HMMs are used to model melody notes, other instrument notes, and silence or noise segments. Key estimation and between-note transition are applied as a musicological model. After transcribing the melody notes, a detailed F0 trajectory is estimated for each note by using the most salient F0 estimates and interpolation with cubic splines.