So, in the prebuilt acoustic models that are available from CMU (and other places), I see the following three lines in the feat.params file:
-transform dct
-round_filters no
-remove_dc yes
I'm trying to train my own models (to be used w/ pocketsphinx), so I started a new model based on the training. The problem is that I can't figure out where these things go. It seems make_feat.pl would be the natural place, but wave2feat does not take these as inputs. -transform is in make_feat.pl, but it isn't actually getting passed to the tool, it just gets stuck out in the feat.params file.
The other two I can't seem to figure out at all.
Help?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
So, in the prebuilt acoustic models that are available from CMU (and other places), I see the following three lines in the feat.params file:
-transform dct
-round_filters no
-remove_dc yes
I'm trying to train my own models (to be used w/ pocketsphinx), so I started a new model based on the training. The problem is that I can't figure out where these things go. It seems make_feat.pl would be the natural place, but wave2feat does not take these as inputs. -transform is in make_feat.pl, but it isn't actually getting passed to the tool, it just gets stuck out in the feat.params file.
The other two I can't seem to figure out at all.
Help?
> It seems make_feat.pl would be the natural place, but wave2feat does not take these as inputs.
If you'll add it to feat.params creation part, It does.