fairseq/tests/speech
Ilia Kulikov 5528b6a382 add reading from zip audio to hubert dataset and scripts (#3403)
Summary:
These are changes from:

https://github.com/fairinternal/fairseq-py/pull/3310
https://github.com/fairinternal/fairseq-py/pull/3285

which were in ust team branch, now moving them to the main.

the main goal is to provide hubert dataset and scripts to read audio from zipped audio storage with backward compatibility depending on the given path.

X-link: https://github.com/fairinternal/fairseq-py/pull/3403

Reviewed By: kahne

Differential Revision: D37150156

Pulled By: uralik

fbshipit-source-id: 7f249b09d7e971c6c7f99114709c26e6a35805cf
2022-06-24 14:09:30 -07:00
..
__init__.py Refactor speech tests and add missing regression tests (#3001) 2022-02-04 14:35:02 -08:00
test_convtransformer_simul_trans.py fix failing convtransformer test (#3107) 2022-02-22 11:24:11 -08:00
test_dual_input_wav_transformer.py Merge STPT: Step 3 2022-05-10 19:44:00 -07:00
test_dualinput_s2t_transformer.py speech integration tests for jointly trained models 2022-01-07 12:45:20 -08:00
test_fastspeech2.py add xm_transformer test; refactor speech tests 2021-12-31 12:31:11 -08:00
test_s2s_transformer.py fix s2s test - disable multitasking by setting multitask_config_yaml to None (#3059) 2022-02-09 10:05:22 -08:00
test_s2t_conformer.py Refactor speech tests and add missing regression tests (#3001) 2022-02-04 14:35:02 -08:00
test_s2t_transformer.py Refactor speech tests and add missing regression tests (#3001) 2022-02-04 14:35:02 -08:00
test_tts_transformer.py add xm_transformer test; refactor speech tests 2021-12-31 12:31:11 -08:00
test_wav2vec2.py Refactor speech tests and add missing regression tests (#3001) 2022-02-04 14:35:02 -08:00
test_xm_transformer.py add reading from zip audio to hubert dataset and scripts (#3403) 2022-06-24 14:09:30 -07:00