i wnt a detailed explanation of pruning parameters(beam, pbeam, wbeam,
subvqbeam etc) of sphinx3 and there effect on phone recognition
accuracy....plzz reply as soon as possible
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
-beam: Determines which HMMs remain active at any given point (frame) during recognition. (Based on the best state score within each HMM.)
-pbeam: Determines which active HMM can transition to its successor in the lexical tree at any point. (Based on the exit state score of the source HMM.)
-wbeam: Determines which words are recognized at any frame during decoding. (Based on the exit state scores of leaf HMMs in the lexical trees.)
-subvqbeam: For each senone and its underlying acoustic model, determines its active mixture components at any frame.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
i wnt a detailed explanation of pruning parameters(beam, pbeam, wbeam,
subvqbeam etc) of sphinx3 and there effect on phone recognition
accuracy....plzz reply as soon as possible
As written in the Hieroglyphs document:
-beam: Determines which HMMs remain active at any given point (frame) during recognition. (Based on the best state score within each HMM.)
-pbeam: Determines which active HMM can transition to its successor in the lexical tree at any point. (Based on the exit state score of the source HMM.)
-wbeam: Determines which words are recognized at any frame during decoding. (Based on the exit state scores of leaf HMMs in the lexical trees.)
-subvqbeam: For each senone and its underlying acoustic model, determines its active mixture components at any frame.