hsercanatli / symbtrsynthesis Goto Github PK

View Code? Open in Web Editor NEW

2.0 2.0 0.0 97.06 MB

The implementations of an adaptive tuning method and a synthesizer for SymbTr scores in MusicXML format.

License: GNU Affero General Public License v3.0

Python 0.03% Jupyter Notebook 99.97%

symbtrsynthesis's People

Contributors

Stargazers

Watchers

symbtrsynthesis's Issues

Licence?

Choose the licence.

arrange symbtr synth I/O

Add MIDI output

Currently the code only outputs audio file. We should also have a MIDI output in the future (whenever Sercan has time)

Convert the functions to classes

carry the inline parameters such as number of harmonics to object initialization etc.

Make sure the code conforms with PEP 8

Convert the current note names to the ones used in SymbTr

Note onsets

Generate note onsets for the synthesized audio

MusicXML reader

Check musicxml reader, also try the one in lilypond converter

peak detection

optimize the peak detection algorithm parameters.

demo script

A demo script would be nice so anybody can use it easily.

I suggest using https://jupyter.org/ (successor ipython notebook) since it´s interactive.

add readme

Write the details

Runtime error in theoretical interval computation

Some of the scores give an error in Dunya:

http://dunya.compmusic.upf.edu/document/manager/collection/makam-symbtr/e25b247d-bb79-4c40-999a-15c822f15a68/75
http://dunya.compmusic.upf.edu/document/manager/collection/makam-symbtr/86e85142-3964-4070-a1a6-59a6606bec26/75
http://dunya.compmusic.upf.edu/document/manager/collection/makam-symbtr/90516943-fb05-4b31-9133-fbbd63dda0e0/75

The error message is:

  Traceback (most recent call last):
  File "/srv/dunya/env/src/pycompmusic/compmusic/extractors/__init__.py", line 137, in process_document
    return self.run(musicbrainzid, fname)
  File "/srv/dunya/env/src/pycompmusic/compmusic/extractors/makam/scoresynthesis.py", line 65, in run
    bpm, measures, tnc_sym, symbtr_slug, mbid)
  File "/srv/dunya/env/src/pycompmusic/compmusic/extractors/makam/scoresynthesis.py", line 81, in synth
    tonic_sym=tnc_sym, synth_type='karplus', verbose=True)
  File "/srv/dunya/env/local/lib/python2.7/site-packages/symbtrsynthesis/adaptivesynthesizer.py", line 100, in synth_from_tuning
    tonic_sym)
  File "/srv/dunya/env/local/lib/python2.7/site-packages/symbtrsynthesis/adaptivesynthesizer.py", line 123, in _replace_tuning
    theo_int = interval_dict[note_sym] - interval_dict[tnc_sym]
KeyError: None

It seems the note_sym somehow becomes None in some cases.

Release a package

make musicxml reader a separate repository

Change the name of the repo from adaptivetuning to symbtrsynthesis

It is more informative and easier to explain acc. to descriptions written in the FMA paper

Use note_model

It will simplify the code greatly by using the note_model repository to find the notes in adaptive tuning

hsercanatli / symbtrsynthesis Goto Github PK

symbtrsynthesis's People

Contributors

Stargazers

Watchers

symbtrsynthesis's Issues

Recommend Projects

Recommend Topics

Recommend Org