AlignmentDuration
AlignmentDuration copied to clipboard
georgid
Lyrics-to-audio-alignement system. Based on Machine Learning Algorithms: Hidden Markov Models with Viterbi forced alignment. The alignment is explicitly aware of durations of musical notes. The phon...
With essentia’s viterbi https://github.com/MTG/essentia/issues/253 or - [Madmom’s Vitebi implementation](https://github.com/CPJKU/madmom/blob/51969404126cf98584ab3fc21efce64ed7342ab3/madmom/ml/hmm.pyx#L434) (cython) - Johan Pawels https://ismir2017.smcnus.org/wp-content/uploads/2017/10/195_Paper.pdf
The LyricsWithModels is not needed for NeuralNEtwork, so Baseclass is used for DNN, add padded silicce method. As result: - LyricsWithModelsBase is used for DNN. LyricsWithModelsCNN._linkTomodels and - LyricsWithModelsBase._linkTomodels do...
onsets.OnsetDetector.OnsetDetector.parseNoteOnsetsGrTruth and replace also parse.TextGrid_Parsing._findBeginEndIndices
sectionLInk object has no section object assigneed. See in method makam.MakamRecording.MakamRecording._loadsectionTimeStampsLinks have a look for an example at: makam.MakamRecording.MakamRecording._loadsectionTimeStampsAnno() This fails in align.LyricsAligner.LyricsAligner.alignRecording: if not hasattr(currSectionLink, 'section') or currSectionLink.section ==...
This workaround https://github.com/georgid/AlignmentDuration/commit/625107b9607feb913681699fffb39130c51027f1 TODO: in the future think of a better idea
in lines starting at if ParametersAlgo.WITH_ORACLE_PHONEMES: # oracle phonemes https://github.com/georgid/AlignmentDuration/blob/noteOnsets/align/LyricsAligner.py#L253
https://github.com/georgid/AlignmentDuration/blob/noteOnsets/src/align/LyricsAligner.py#L142 test from for_makam/lyrics_align WITH_DURATIONS = True first - get rid of loading htk models, and get rid of class LyricsWithModelsHTK (see issue https://github.com/georgid/AlignmentDuration/issues/40) - select recordings with no second...
Copy the essentia c++ code for extraction of MFCC in repository. This way the dependency on essentia will be not needed. MFCC extraction is done in [FeatureExtractor._extractMFCC](https://github.com/georgid/AlignmentDuration/blob/production_no_melody/src/align/FeatureExtractor.py#L136)