We are using CMUCLMTK for Language Model. We also tried SRILM but accuracy is going down if we compare it with CMUCLMTK. Which language model toolkit we need to use for increasing the accuracy. We are using all this with 18 hours of transcripted text.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We are using CMUCLMTK for Language Model. We also tried SRILM but accuracy is going down if we compare it with CMUCLMTK. Which language model toolkit we need to use for increasing the accuracy. We are using all this with 18 hours of transcripted text.
You need to compare perplexity, not accuracy. srilm accuracy should be better given you use proper options.