View Code? Open in Web Editor
NEW
Code for the paper: Audio to Score Matching by Combining Phonetic and Duration Information
Home Page: https://arxiv.org/abs/1707.03547
License: GNU Affero General Public License v3.0
Python 85.08%
C 0.36%
MATLAB 14.57%
jingjusingingphrasematching's Issues
A question:
I would like to reuse your code for expansion from a syllable in pinyin to the labels of the classes of the CNN model (based on x-sampa).
I saw that code for creating a transition matrix is essentially what I need, right?
I dont see though why you have only a list of the syllable finas expanded to x-sampa? What happends to the initials? Are they omitted because they are too short or s.th.?