Florian Metze
Maybe something to do with the three-state vs single-state topology? I must admit I am not familiar with these Kaldi tools - sorry
Yes, the tf_clean branch has all the code, we are cleaning up a few things currently. @ramonsanabria has more details
I have not tried an RNNLM with Chinese characters, but with a large enough training corpus, you should be able to directly use chinese characters, not? Isn’t this also what...
pull request would be great, please! > On Sep 27, 2017, at 6:17 AM, Xingyu Na wrote: > > https://help.github.com/articles/about-pull-requests/ > > 2017-09-27 17:10 GMT+08:00 Sundy1219 : > > >...
Thanks, does this cause any trouble? Do you know how to fix it?
it depends. you can start from any of the existing data preparation scripts. look at the local/prepare_data.sh scripts in the individual recipes, and see which one you can most easily...
Some stability problem with file access? Are these local files, or being accessed via NFS or some other shared file system? Always for process 1, i.e. the master thread?
if the fix worked, would you mind submitting it? thanks!
Has all of this been tried over NFS file systems, or local discs? I have not worked with this code in a long time, but I think the best way...
It is typically a good strategy to copy all data that you are going to need onto the VM anyway at the beginning of the training process, because access will...