SPRAAK
|
spr_add_gauss | Increase the level of tying in a reduced SC_HMM |
spr_addkey | Add/modify a key in the header of a KEY file |
spr_bmix | mix the data in a file |
spr_calc_lda | calculate the lda transformation for a given preprocessing and |
spr_calc_oprobs | Dump state output probs |
spr_copy | File copying with conversion |
spr_ct2hmm | Combine the statistics from multiple segpass or vitpass runs into an updated model. |
spr_cwr_main | Large Vocabulary Continuous Speech Recognition with Hidden Markov models |
spr_decorr | Calculate a good decorrelation matrix given the full covariance |
spr_delkey | Delete a key in the header of a KEY file |
spr_fv_gauss | feature space MLLR |
spr_fwdbwd_probs | Fwd-bwd alignment (viterbi alike) te determine boundary probabilities |
spr_getkey | Get a key from the header of a KEY file |
spr_hmmc2d | Convert a (semi)continuous density HMM to a discrete HMM or to a not-reduced semi-continuous HMM |
spr_hmmcheck | Check the validity of an HMM |
spr_hmmd2c | Convert a discrete HMM to a continuous or a semi-continuous HMM |
spr_hmmmerge | Concatenate two HMMs |
spr_hmmnormalize | Normalizes input file so that probabilities sum up to one |
spr_hmmprint | Print the values in an HMM file |
spr_hmmshake | Reorder the states in an HMM |
spr_hmmsmooth | Smooth the mixture weights in HMM's retrained on little data |
spr_hmmsum | Combine two HMMs by summing the moments |
spr_htk2key | Convert from HTK format to the KEY format |
spr_kappend | Append KEY files |
spr_kdiff | Compare KEY files |
spr_knn_dtw | find the k nearest neighbour templates using DTW for a given context |
spr_lex_cvt | Modify a lexicon on a line-by-line basis. |
spr_lm_arpabo | Convert arpa style N-gram's to and from the binary format used in SPRAAK |
spr_mdt_make_cgt | create cluster gaussian to back-end gaussian association table |
spr_msg_exec | execute + time a program |
spr_mvg_init | Initialization for K-means cluster centra and multi variate gaussians |
spr_printf | test program |
spr_scoreres | (Re)calculate the CSR score for a given result file. |
spr_segpass | Initialize an hmm file on the basis of unit or state level segmentations |
spr_sel_frames | Select frames corresponding to units or states, or to all data |
spr_sel_gauss | Reduce the number of Gaussians in a mixture |
spr_setminprob | Set minimal probabilities in an HMM. |
spr_sigp | signal processing |
spr_space_cake | Continue Automatische Klank-Evaluatie |
spr_spraak_socket | Speech interface using sockets |
spr_tree2model | Construction of acoustic models given a decision tree |
spr_tree_create | Creation of phonetic decision trees for context clustering |
spr_tree_gauss_split | Decision tree based splitting of gaussians for HMMs with tied gaussians |
spr_tt | test program |
spr_vitalign | Viterbi alignment. |
spr_vitpass | one pass of viterbi training for HMM-models |
spr_vitprobs | Viterbi alignment + path probabilities |
spr_wlatp | processing of word lattices |