Re: [Kaldi-users] Memory requirement for FSTs

[Andreas S. K. <as...@cb...>]

Hi E 

I have successfully trained and used 40k language models and run into the same problem as you do when trying to increase the vocab. Because of Danish productive compounding, I have decided try the 'lmrescore' approach that Dan suggested in an earlier mail.

-Andreas
________________________________________
From: E [oth...@ao...]
Sent: 18 October 2013 06:06
To: kal...@li...
Subject: Re: [Kaldi-users] Memory requirement for FSTs

Thanks a lot for the answers and possible solutions.

Few questions-

What is the maximum size trigram language model supported by FST? I tried to use Gigaword LM (64k vocab), mkgraph.sh ran for a long time but crashed afterwords. So I want to know if there is a theoretical limit on size of language model that can be integrated with Kaldi.

I will try to make HCLG.fst with gigaword again (with triphone AM), but has anyone tried to build it with LM of this size successfully, if so, what were the system requirements (RAM) and final FST size in mega/gigabytes?