Граф коммитов

59 Коммитов

Автор SHA1 Сообщение Дата
David Snyder 4ed9fc8c17 sandbox/language_id: In run.sh now combines and uses all sre08 training data
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3573 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-23 17:39:58 +00:00
David Snyder 5e2f176bdc sandbox/language_id: Adding a needed sort checks in utils/fix_data_dirs.sh
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3560 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-21 04:15:56 +00:00
Nagendra Goel cd4482adb0 add missing file
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3555 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-20 19:55:31 +00:00
Nagendra Goel e4debea29c fix filename
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3554 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-20 18:16:01 +00:00
David Snyder 8bd7157ffd sandbox/language_id: Adding an incomplete run.sh file for this example.
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3548 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-20 06:35:42 +00:00
David Snyder df7f33d7df sandbox/language_id: Adding scripts for vad decisions, ubm training and ivector extraction for language_id. These scripts are copied from the speaker_id setup, with modifications. For instance, we use SDC features instead of delta-deltas, and lang2utt instead of spk2utt.
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3547 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-20 06:26:47 +00:00
David Snyder 18929a54a9 sandbox/language_id: fixes to local/make-sre-2008-train.pl and utils/fix_data_dir.sh for language-id.
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3544 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-20 05:59:03 +00:00
Nagendra Goel 4c191d7c98 fix data prep
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3542 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-19 18:37:22 +00:00
David Snyder d382fc9ba5 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.
git-svn-id: https://svn.code.sf.net/p/kaldi/code/sandbox/language_id@3533 5e6a8d80-dfce-4ca6-a32a-6e07a63d50c8
2014-02-18 01:05:05 +00:00