I see that the parameters of Bavieca are the following:
configuration_file
hypotheses_file
batch_file
So configuration_file is EE.cfg, and hypotheses_file must be the output directory of the decoded data right? What about batch_file; is that where the WSJ data is specified?
Thank you,
Matt
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello Matt. You got it right, the batch file is where you pass the WSJ test data to generate the hypotheses, which will be stored into the hypotheses file. You can find information about the format of the batch file in www.bavieca.org
Dani
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello Daniel,
I am trying to run a simple test on wsj and am having trouble understanding the command-line parameters of Bavieca.
Here is what I have:
configurationFile = config/EE.cfg
phoneticSymbolSet.file = test/20k/languageModel/CMU/phoneset.txt
acousticModels.file = data/bavieca/wsj_models.bin
languageModel.file = test/20k/languageModel/wsj-20k-onp-trigram.arpa
lexicon.file = test/20k/languageModel/wsj-20k.lex
wsjdata = data/wsj
I see that the parameters of Bavieca are the following:
configuration_file
hypotheses_file
batch_file
So configuration_file is EE.cfg, and hypotheses_file must be the output directory of the decoded data right? What about batch_file; is that where the WSJ data is specified?
Thank you,
Matt
Hello Matt. You got it right, the batch file is where you pass the WSJ test data to generate the hypotheses, which will be stored into the hypotheses file. You can find information about the format of the batch file in www.bavieca.org
Dani