This is the accompanying webpage for the paper:
Zhiyao Duan and Bryan Pardo, A state space model for online polyphonic audio-score alignment, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2011, 197-200. <pdf> <poster>
As described in the paper, we compare three audio-score alignment methods:
- SF-Pitch: the proposed score follower using the multi-pitch observation model.
- SF-Chroma: the proposed score follower using the chroma observation model.
- Scorealign: an open source software, based on the algorithm described in [1].
Here we show four examples, one for each polyphony. We show the results by time-warping the original mixture according to the output alignment of each score follower/aligner. You may follow the following steps to get a feel for these examples:
- Import the score file (*.mid) to Audacity;
- Import the time-warped audio file (*.wav) to Audacity;
- Listen to the audio and watch how well it is aligned with the score.
| Polyphony | Score | Original audio | Time-warped audio by SF-Pitch | Time-warped audio by SF-Chroma | Time-warped audio by Scorealign |
|---|---|---|---|---|---|
| 1 | MIDI | wave | wave | wave | wave |
| 2 | MIDI | wave | wave | wave | wave |
| 3 | MIDI | wave | wave | wave | wave |
| 4 | MIDI | wave | wave | wave | wave |
References
[1] N. Hu, R.B. Dannenberg and G. Tzanetakis, "Polyphonic audio matching and alignment for music retrieval," in Proc. 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), New Paltz, New York, USA, 2003, pp. 185-188.
For any questions or comments, please contact us at zhiyaoduan00 AT gmail DOT com.