I have trained a model and used it to force align certain sentences. About 60% of the words are getting aligned. Now I want to specify wider beamwidth for alignment so it rejects fewer words (at the cost of accuracy I guess). Is there a way to specify the beamwidth on a model that is already built? Or to specify it in the SpeechAligner (edu.cmu.sphinx.api.SpeechAligner) instantiation?
I know we can specify it while building model with $CFG_FORCE_ALIGN_BEAM but I am not sure if that sticks around somehwere in the model definition once it is built.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I know we can specify it while building model with $CFG_FORCE_ALIGN_BEAM but I am not sure if that sticks around somehwere in the model definition once it is built.
No, it is only used on sphinx3_align tool invocation.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have trained a model and used it to force align certain sentences. About 60% of the words are getting aligned. Now I want to specify wider beamwidth for alignment so it rejects fewer words (at the cost of accuracy I guess). Is there a way to specify the beamwidth on a model that is already built? Or to specify it in the SpeechAligner (edu.cmu.sphinx.api.SpeechAligner) instantiation?
I know we can specify it while building model with $CFG_FORCE_ALIGN_BEAM but I am not sure if that sticks around somehwere in the model definition once it is built.
No, it is only used on sphinx3_align tool invocation.