kaldi

История

Dan Povey 6f598676cc several nnet2-online changes: make it easier to get the feature extraction options right in cross-system training; add train_pnorm_simple.sh script (simplified learning-rate schedule and improved combination at the end, supersedes train_pnorm_fast.sh); modifying big-data online-nnet2 recipes to use 40-dimensional MFCC rather than 13 as input (will add results soon, but they are improved). Modified filter_scp.pl to have one-based, not zero-based, field index. git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@4493 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8		2014-09-30 19:18:36 +00:00
..
v1	several nnet2-online changes: make it easier to get the feature extraction options right in cross-system training; add train_pnorm_simple.sh script (simplified learning-rate schedule and improved combination at the end, supersedes train_pnorm_fast.sh); modifying big-data online-nnet2 recipes to use 40-dimensional MFCC rather than 13 as input (will add results soon, but they are improved). Modified filter_scp.pl to have one-based, not zero-based, field index.	2014-09-30 19:18:36 +00:00
README.txt	sandbox/language_id: Creating example directory in /egs/lre/v1 with common and shared (linked with wsj) scripts. Also adding local/make-sre08-data.sh which creates the training data for this example.	2014-02-18 01:05:05 +00:00

README.txt

 This directory (lre) will contain example scripts for language identification.
 Development is just beginning so this recipe is incomplete.
 The following LDC corpora are required:
    
   SRE 2008 training set:  LDC2011S05 
 
 This list will be updated as scripts for system development and testing
 (which will require additional data sources) are created. 

 The subdirectories "v1" and so on are different versions of the recipe.