Sequential modeling

librosa.sequence provides functions for sequential processing, such as dynamic time warping (DTW), recurrence quantification analysis (RQA), and various forms of Viterbi decoding.

Helper utilities are also provided for constructing transition matrices or post-processing alignment paths.

dtw	Dynamic time warping (DTW).
rqa	Recurrence quantification analysis (RQA)
path_to_steps	Convert a DTW warping path to an array of fractional steps.
viterbi	Viterbi decoding from observation likelihoods.
viterbi_discriminative	Viterbi decoding from discriminative state predictions.
viterbi_binary	Viterbi decoding from binary (multi-label), discriminative state predictions.
transition_uniform	Construct a uniform transition matrix over n_states.
transition_loop	Construct a self-loop transition matrix over n_states.
transition_cycle	Construct a cyclic transition matrix over n_states.
transition_local	Construct a localized transition matrix.