Vahid Noroozi cfcf694e30 Adding parallel transcribe for ASR models - suppports multi-gpu/multi-node (#3017) ... * added transcribe_speech_parallel.py. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed style. Signed-off-by: Vahid <vnoroozi@nvidia.com> * removed comments. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed bug. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed bug. Signed-off-by: Vahid <vnoroozi@nvidia.com> * added comments inside the script. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed speed_collate_fn for TTS. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed speed_collate_fn for TTS. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed speed_collate_fn for TTS. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed speed_collate_fn for TTS. Signed-off-by: Vahid <vnoroozi@nvidia.com> * added return_sample_id. Signed-off-by: Vahid <vnoroozi@nvidia.com> * added return_sample_id. Signed-off-by: Vahid <vnoroozi@nvidia.com> * merged dataset configs. Signed-off-by: Vahid <vnoroozi@nvidia.com> * merged dataset configs. Signed-off-by: Vahid <vnoroozi@nvidia.com> * dropped sample_ids from train and validation. Signed-off-by: Vahid <vnoroozi@nvidia.com> * dropped sample_ids from train and validation. Signed-off-by: Vahid <vnoroozi@nvidia.com> * reverted tts patches. Signed-off-by: Vahid <vnoroozi@nvidia.com> * reverted tts patches. Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed the default values in the dataset's config Signed-off-by: Vahid <vnoroozi@nvidia.com> * fixed the default values in the dataset's config Signed-off-by: Vahid <vnoroozi@nvidia.com> * Fixed the bug for optional outputs. Signed-off-by: Vahid <vnoroozi@nvidia.com> * Fixed the bug for optional outputs. Signed-off-by: Vahid <vnoroozi@nvidia.com> * addressed some comments. Signed-off-by: Vahid <vnoroozi@nvidia.com> * disabled dali support for return_sample_id. Signed-off-by: Vahid <vnoroozi@nvidia.com> * disabled dali support for return_sample_id. Signed-off-by: Vahid <vnoroozi@nvidia.com> * converted the config to omegaconf. Signed-off-by: Vahid <vnoroozi@nvidia.com> * converted the config to omegaconf. Signed-off-by: Vahid <vnoroozi@nvidia.com> * converted the config to omegaconf. Signed-off-by: Vahid <vnoroozi@nvidia.com> * moved wer/cer calculation to the end. Signed-off-by: Vahid <vnoroozi@nvidia.com> * moved wer/cer calculation to the end. Signed-off-by: Vahid <vnoroozi@nvidia.com> * moved wer/cer calculation to the end. Signed-off-by: Vahid <vnoroozi@nvidia.com> * moved wer/cer calculation to the end. Signed-off-by: Vahid <vnoroozi@nvidia.com> * moved wer/cer calculation to the end. Signed-off-by: Vahid <vnoroozi@nvidia.com> * calculates global wer instead of per sample. Signed-off-by: Vahid <vnoroozi@nvidia.com> * calculates global wer instead of per sample. Signed-off-by: Vahid <vnoroozi@nvidia.com> * calculates global wer instead of per sample. Signed-off-by: Vahid <vnoroozi@nvidia.com>		2021-11-10 00:37:19 -08:00
..
collections	Adding parallel transcribe for ASR models - suppports multi-gpu/multi-node (#3017)	2021-11-10 00:37:19 -08:00
core	Merge r1.5.0 bugfixes and doc updates to main (#3133)	2021-11-04 10:26:58 -06:00
core_ptl	Remove PTL 1.4 upper bound (#2600)	2021-08-10 12:50:22 -07:00
hydra	Fix hydra runner (#1652)	2021-01-22 10:26:46 -05:00
nemo_text_processing	Itn fr (#2921)	2021-09-30 14:00:28 -07:00
__init__.py	ITN Ru and non-deterministic TN (#2519)	2021-08-12 22:00:35 -07:00
check_copyright_header.py	Itn german (#2486)	2021-07-20 12:41:36 -07:00
conftest.py	Update conftest.py to auto untar local data (#2655)	2021-08-13 14:03:13 -07:00
manualtest_model_downloads.py	Update model names (#2845)	2021-09-19 18:58:04 -07:00
py_cprheader.txt	Add copyright headers check (#2490)	2021-07-15 15:06:45 -04:00
test_data_dir.py	Test data dir (#951)	2020-07-30 09:45:40 -07:00