Georgi Dzhambazov

icon indicating a website link https://www.linkedin.com/in/georgi-d-/ icon indicating an email [email protected]

icon company Freelance icon location Sofia icon location I bring technology closer to the society by creating innovative, AI-enabled, software solutions - mainly in the domains of music, speech and natural language

Results 55 issues of Georgi Dzhambazov

improve roganization of MusicXML Parser: use ScoreSection class in MusicXMLParser

Would be nice that the field sectionLyrics in MUsicXML Parser is type ScoreSection. further MakamScore should have a field score in the same way symbTrParser has a field score of...

reduce code of constructTransMAtrix

in function constructTransMAtrix from Decoder : # SPECIAL CASE: two last states the else is not needed? because defineForwardTransProbs() covers it. Related: THink how to reduce code do that #...

- put reading pitch as input in lyricsalign not in align.FeatureExtractor.loadMFCCs

enhancement

when withPaddedSilence, consider creating dummy Word with one phoneme sp

in align.LyricsWithModels.LyricsWithModels._linkToModels This way not needed to handle specially the phoneme sp. It is now inserted as Phoneme in phonemes network in LyricsWithModels but not seen in list of words...

enhancement

refine and commit LyricAligner

-LyricsAligner. withAnnotations and withLinks have two similar loops, think how to put it in one loop. - test and update pycompmusic.lyricsAlign

hmm.Path.Path.

hmm.Path.Path.**init** can haas input only hmm and read psi and phi from it

dont trim file,

1 comment

so eliminate RecordingSegmenter step, do alignment for each section in a loop - recomupte timestamps fron whole recording mfc to given sectuin( input: section.json) - Align - recomupte tinmestamps for...

optimize: do observation prob for a given feature vector for all phonemes in alphabet, not for all phonemes in phonemeNetwork

clean code in function doitOneChunk.decodeAudioChunk()

# TODO: could be done weasier with this code, and check last method in Word # grTruthWordList = testT(decoder.lyricsWithModels)

enhancement

REDUCE CODE: check if we can replace some of the code in SymbTrPrser.syllable2LyricsOneSection with code from LyricsParsing.expandlyrics2WordList.

as well LyricsWithModels.LyricsWithModels.printWordsAndStates and LyricsWithModels.LyricsWithModels.printWordsAndStatesAndDurations can take code from LyricsParsing.expandlyrics2WordList.

enhancement