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 phonetic model are classified with MLP Deep Neural Network.
in function detectedAlignedfileName = mlf2TabFormat(detectedWordList, URIrecordingNoExt, tokenLevelAlignedSuffix)