I have already trained a system on english corpus containing 13K sentences,
now when I am decoding my test of 198 sentences it is taking around two days
to decode. I am using a server PC so it should not take that much time right??
My training was done in around 6hrs but while training force allignment was
off so few errors were give so it can be the reason?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It seems that the reason for slow decoding is inaccurately trained acoustic
model. Your model should be reasonable quality in order to allow fast
decoding.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have already trained a system on english corpus containing 13K sentences,
now when I am decoding my test of 198 sentences it is taking around two days
to decode. I am using a server PC so it should not take that much time right??
My training was done in around 6hrs but while training force allignment was
off so few errors were give so it can be the reason?
Looking on your other topic
https://sourceforge.net/projects/cmusphinx/forums/forum/5471/topic/5218747?me
ssage=11418334
It seems that the reason for slow decoding is inaccurately trained acoustic
model. Your model should be reasonable quality in order to allow fast
decoding.