Adapt Kaldi-ASR nnet3 chain models from Zamia-Speech.org to a different language model.
Constructive comments, patches and pull-requests are very welcome.
Tutorial
To create the language model we would like to adapt our kaldi model to, we first need to create a set of sentences. To get started, download and uncompress a generic set of sentences for you language, e.g.
now suppose the file utts.txt contained the sentences you would like the model to recognize with a higher probability than the rest. To achieve that, we add these sentences five times in this examples to our text body:
Now we can start the kaldi model adaptation process:
kaldi-adapt-lm ${MODEL} lm.arpa mymodel
You should now be able to find a tarball of the resulting model inside the work subdirectory.
If at the end of adaptation process you have a lot of messages like "cp: cannot stat 'exp/adapt/graph/HCLG.fst': No such file or directory", then highly likely you run out of memory during adaptation process. (For example adapting kaldi-generic-en-tdnn_250 model consumes near 12Gb of RAM)