git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5224 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5223 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

- 'blocksoftmax' is better name as it has same name as the used output component,
- we use multitask objective function, so we can see performance on individual tasks,
- also adding some older LSTM results on tedlium, not working on this recipe now,



git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5222 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5219 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5218 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5217 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

trunk: adding optional 3rd argument to lattice-to-ctm-conf so that it can also reads the word sequence from input

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5215 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5214 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5212 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

(trunk) Patch the rnnlm code to allow longer filenames + add a error message when the filename is still longer than the buffer size

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5210 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5205 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5204 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

(trunk) The decode scripts now fail when the scoring fails. Added --skip-scoring switch to all of them to allow skipping the scoring alltogether, i.e. to avoid the decoding failure

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5202 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

trunk,nnet1 : bugfix in cuda-gpu detection (mmi/mpe script itself is not running on a machinge with GPU, it submits job to a GPU node)


git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5201 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5197 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5196 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5195 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5193 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

trunk: Adding binaries fgmm-global-acc-stats-post and fgmm-global-init-from-accs which will be used primarily for DNN-SID scripts.

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5192 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5191 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5189 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5187 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

(trunk) Adding compute-atwv -- a tool to (fast) computation of the ATWV and the other TWV metrics + some other negligible fixes

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5186 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

- it works, but it needs further analysis, so far it did not outperform the LSTM.



git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5183 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5182 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5181 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5180 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5178 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5177 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

(trunk) Create a separate kws library. If you are using lat/kaldi-kws.h, change it to kws/kaldi-kws.h.

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5176 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5175 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5174 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5172 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

trunk,nnet1 : 
- ran cpplint.py on nnet-convolutional-2d-component.h, nnet-max-pooling-2d-component.h, nnet-average-pooling-2d-component.h, to make it compliant w.r.t whitespaces



git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5171 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5170 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5169 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

- faster cnn2d by pre-computing scaling factors used on gradients of cnn2d weights
- unit tests for all methods of cnn2d, max-pool-2d, avg-pool-2d components
- added cnn2d recipe in Resource Management, local/nnet/run_cnn2d.sh
bugfix steps/nnet/make_priors.sh : in importing cmvn-option in preparing feature-extraction pipeline 
 


git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5165 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5163 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5156 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8

(trunk) Fix issue when the lines wouldn't be random enough. WARNING: this patch will cause shuffle_list.pl working in different way, i.e. it will produce differently shuffled lines! 

git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@5153 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8