A COMPARISON OF PROBABILISTIC MODELS FOR ONLINE PITCH TRACKING

Publication Type:

Conference Paper

Source:

SMC Conference 2010 (2010)

URL:

files/proceedings/2010/74.pdf

Abstract:

In this study, we propose and compare two probabilistic models for online pitch tracking: Hidden Markov Model and Change Point Model. In our models each note has a certain characteristic spectral shape which we call spec- tral templates. Hence the system’s goal is to find the note whose template is active given the audio data. The main focus on this work is the trade off between latency and ac- curacy of the pitch tracking system. We present the prob- abilistic models and the inference schemes in detail. En- couraging results are obtained from the experiments that are done on low-pitched monophonic audio.