14 Коммитов

Автор	SHA1	Сообщение	Дата
David Snyder	f4fd8dd76b	bn-music-speech-fix: Update to run.sh to use the LDC distribution of the HUB4 corpus, update to local/make_bn.py so that it is more flexible with regard to differences in source directory format	2016-04-12 18:25:31 -04:00
David Snyder	eb5b32cf78	bn-music-speech-fix: Changing LDC97S66 to LDC97S44 in corpra list for bn-music-speech.	2016-04-12 15:22:17 -04:00
David Snyder	fb01718248	bn-music-speech-fix: Changing LDC numbers in egs/bn_music_speech to the correct corpra.	2016-04-12 14:17:59 -04:00
Jan Trmal	12539e86cd	moving the src/path.sh into tools/config/common_path.sh	2016-03-30 09:57:03 -04:00
Jan Trmal	9c0391c1ca	all other path.sh fixed	2016-03-26 23:46:43 -04:00
Dan Povey	b3bbc03868	modifying cmd.sh in example recipes to encourage the use of new-style queue options and conf/queue.conf	2016-02-18 16:58:46 -05:00
David Snyder	0eec30a9f9	music-speech: Adding symbolic link to the sid directory in egs/bn_music_speech/v1	2015-12-16 10:51:21 -05:00
David Snyder	bb870e8a46	music-speech: Fixing bug in binary merge-vads and in local/make_musan.py for egs/bn_music_speech	2015-12-09 13:42:15 -05:00
David Snyder	69073037f4	music-speech: Improving documentation, style, and config handling in binaries compute-vad-from-frame-likes and merge-vads. In egs/bn_music_speech: removing old code from cmd.sh, replacing conf/merge_vad.conf with conf/merge_vad_map.txt (and adding comments to it), changing how python scripts are invoked.	2015-12-04 17:34:08 -05:00
David Snyder	cf07dfd703	music-speech: Adding config file needed for bn example	2015-12-02 16:40:23 -05:00
David Snyder	a3c3383511	music-speech: minor fix to run.sh	2015-11-24 15:01:14 -05:00
David Snyder	7e9e6661af	music-speech: Adding sid/compute_vad_decision_gmm.sh, a script that uses GMM likelihoods to compute VAD decisions	2015-11-24 14:58:23 -05:00
David Snyder	6d688456b9	music-speech: Cosmetic fix	2015-11-22 00:03:08 -05:00
David Snyder	ba32df6d89	Adding bn_music_speech example which demonstrates music/speech discrimination using the MUSAN corpus and Broadcast news. Also adding binaries compute-vad-from-frame-likes and merge-vads that will be used for frame-level GMM-based VAD.	2015-11-21 23:09:18 -05:00