kaldi

История

Jan Trmal a5ecb5839d (trunk) Fixing the scripts using the mem_free/ram_free SGE resource requests to have ram_free on per-job basis instead of per-thread basis. This is in accordance with the changes in kaldi docs git-svn-id: https://svn.code.sf.net/p/kaldi/code/trunk@4753 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8		2015-01-06 11:26:22 +00:00
..
ami	trunk: remove tiedbin in path.sh	2014-10-22 19:30:45 +00:00
aurora4/s5	Updates to various data preparation scripts so validation checks on 'lang' directories will pass. It's possible some of these changes will break some setups, but it's not feasible to fully test this right now.	2015-01-02 03:38:04 +00:00
babel	(trunk) Fixing the scripts using the mem_free/ram_free SGE resource requests to have ram_free on per-job basis instead of per-thread basis. This is in accordance with the changes in kaldi docs	2015-01-06 11:26:22 +00:00
callhome_egyptian	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
chime_wsj0	Updates to various data preparation scripts so validation checks on 'lang' directories will pass. It's possible some of these changes will break some setups, but it's not feasible to fully test this right now.	2015-01-02 03:38:04 +00:00
farsdat	trunk: various changes that relate to G.fst ilabel sorting.	2014-11-15 04:03:49 +00:00
fisher_callhome_spanish	(trunk) Fixing the scripts using the mem_free/ram_free SGE resource requests to have ram_free on per-job basis instead of per-thread basis. This is in accordance with the changes in kaldi docs	2015-01-06 11:26:22 +00:00
fisher_english	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
fisher_swbd/s5	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
gale_arabic	Add gale_mandarin setup; fix a few things in the gale_arabic setup	2014-10-25 02:45:25 +00:00
gale_mandarin/s5	Add gale_mandarin setup; fix a few things in the gale_arabic setup	2014-10-25 02:45:25 +00:00
gp	trunk: remove tiedbin in path.sh	2014-10-22 19:30:45 +00:00
hkust	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
librispeech	Updates to various data preparation scripts so validation checks on 'lang' directories will pass. It's possible some of these changes will break some setups, but it's not feasible to fully test this right now.	2015-01-02 03:38:04 +00:00
lre	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
lre07	trunk: Updating SID and LID scripts to reflect change in how mem_free and ram_free parameters are used for multithreaded scripts on the CLSP grid.	2015-01-06 04:28:29 +00:00
rm	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
sprakbanken	Updates to various data preparation scripts so validation checks on 'lang' directories will pass. It's possible some of these changes will break some setups, but it's not feasible to fully test this right now.	2015-01-02 03:38:04 +00:00
sre08	trunk: Updating SID and LID scripts to reflect change in how mem_free and ram_free parameters are used for multithreaded scripts on the CLSP grid.	2015-01-06 04:28:29 +00:00
swbd	trunk,nnet1 : adding the tandem recipe,	2015-01-02 21:05:51 +00:00
tedlium/s5	trunk: minor fix to prepare_dict script for tedlium, to pass current validation script (remove dups)	2015-01-05 23:22:49 +00:00
tidigits	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
timit	(trunk) Fixing the scripts using the mem_free/ram_free SGE resource requests to have ram_free on per-job basis instead of per-thread basis. This is in accordance with the changes in kaldi docs	2015-01-06 11:26:22 +00:00
voxforge	trunk: minor fix to voxforge script to work on mac.	2014-12-30 04:50:01 +00:00
vystadial_cz	trunk: remove tiedbin in path.sh	2014-10-22 19:30:45 +00:00
vystadial_en	trunk: merging sandbox/dan back to trunk. Includes addition of recipe for the LibriSpeech corpus, and the capability to rescore lattices using ARPA language models that are too big to convert into FSTs.	2014-10-04 18:19:15 +00:00
wsj	(trunk) Fixing the scripts using the mem_free/ram_free SGE resource requests to have ram_free on per-job basis instead of per-thread basis. This is in accordance with the changes in kaldi docs	2015-01-06 11:26:22 +00:00
yesno	trunk: deleting topo.proto files (no longer used since a long-ago script change)	2014-12-31 01:51:48 +00:00
README.txt	trunk: minor, cosmetic changes	2014-03-30 17:53:53 +00:00

README.txt

This directory contains example scripts that demonstrate how to 
use Kaldi.  Each subdirectory corresponds to a corpus that we have
example scripts for.

Note: we now have some scripts using free data, including voxforge,
vystadial_{cz,en} and yesno.  Most of the others are available from
the Linguistic Data Consortium (LDC), which requires money (unless you
have a membership).

If you have an LDC membership, probably rm/s5 or wsj/s5 should be your first
choice to try out the scripts.