Estimate the fundamental frequency F0 and inharmonicity coefficient B of an isolated piano note
The code implements the estimation approach in
F. Rigaud, B. David, and L. Daudet,
“A parametric model and estimation techniques for the inharmonicity and tuning of the piano” ,
The Journal of the Acoustical Society of America, vol. 133, no. 5, pp. 3017–3118, 2013.
The paper describes a robust new algorithm based on the Non-negative Matrix Factorization (NMF) frameworks in order to finely estimate (F0,B) from isolated notes.
- Clone the directory
$ git clone https://github.com/beiciliang/estimate-f0-inharmonicity.git
- Install the requirements using pip
$ cd estimate-f0-inharmonicity
$ pip install --user -r requirements.txt
- Run the python file
$ python estimate-f0-inharmonicity.py [your path containing .wav file of an isolated piano note] [The midi number of the note in the given wav file]
using the data provided in this repo as an example:
$ python estimate-f0-inharmonicity.py ./data/IS-v96-m60.wav 60
it will return:
For note MIDI-No.60:
the estimated fundamental frequency is 262.399515937
the estimated inharmonicity coefficient is 0.000326411619295
The source code could also calculate the first 30th partials and their amplitude based on the estimated (F0,B). Feel free to use it if it's useful for your piano-related projects.
P.S. A matlab implementation is provided at Tian Cheng's soundsoftware repository.