-
Notifications
You must be signed in to change notification settings - Fork 2.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
NFA updates #6695
NFA updates #6695
Commits on Mar 11, 2023
-
update V_NEGATIVE_NUM constant to make better use of torch.float32 range
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 037194b - Browse repository at this point
Copy the full SHA 037194bView commit details -
adjust backpointers dtype if U_max too large
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for cb69ccc - Browse repository at this point
Copy the full SHA cb69cccView commit details -
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0dd8729 - Browse repository at this point
Copy the full SHA 0dd8729View commit details -
Remove need for user to specify model_downsample_factor
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for cceab7c - Browse repository at this point
Copy the full SHA cceab7cView commit details
Commits on Mar 13, 2023
-
change model.cfg.sample_rate to model.cfg.preprocessor.sample_rate
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for f9489d8 - Browse repository at this point
Copy the full SHA f9489d8View commit details -
add check to make sure that window_stride is in model.cfg.preprocessor
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 7b37972 - Browse repository at this point
Copy the full SHA 7b37972View commit details
Commits on Mar 15, 2023
-
reduce memory consumption of backpointers by making them relative ins…
…tead of absolute Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0cef35e - Browse repository at this point
Copy the full SHA 0cef35eView commit details -
update librosa.get_duration() 'filename' param to 'path'
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 11f3430 - Browse repository at this point
Copy the full SHA 11f3430View commit details
Commits on Mar 16, 2023
-
Do not throw error if 'text' or 'pred_text' are empty and make sure C…
…TM filepaths in the output manifest are null Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 9d9b7b2 - Browse repository at this point
Copy the full SHA 9d9b7b2View commit details -
preprocess input text by removing any duplicate spaces and converting…
… any newlines to spaces Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for d916db4 - Browse repository at this point
Copy the full SHA d916db4View commit details
Commits on Apr 4, 2023
-
Use Utterance dataclass instead of dictionaries for keeping track of …
…token/word/segment alignments Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 643a8ee - Browse repository at this point
Copy the full SHA 643a8eeView commit details -
Merge branch 'main' into nfa_updates
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 2be92bf - Browse repository at this point
Copy the full SHA 2be92bfView commit details
Commits on Apr 5, 2023
-
refactor so can save alignments as ctm and ass format files
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0897f33 - Browse repository at this point
Copy the full SHA 0897f33View commit details -
fix bugs for saving character based ASS files and for using pred_text…
… to do alignment Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for c74a63f - Browse repository at this point
Copy the full SHA c74a63fView commit details
Commits on Apr 6, 2023
-
Make token level .ass file use tokens with recovered capitalization
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for f7c920e - Browse repository at this point
Copy the full SHA f7c920eView commit details -
Do not try to generate alignment files if text or pred text is empty,…
… or if number of tokens is too large for T Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 45e3fb1 - Browse repository at this point
Copy the full SHA 45e3fb1View commit details -
rename output manifest file to say '_with_output_file_paths.json'
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for a57409d - Browse repository at this point
Copy the full SHA a57409dView commit details
Commits on Apr 7, 2023
-
add flag to resegment ass subtitle file to fill available text space
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for dbd5232 - Browse repository at this point
Copy the full SHA dbd5232View commit details
Commits on Apr 8, 2023
-
Fix bug in resegmentation code
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 10f85e3 - Browse repository at this point
Copy the full SHA 10f85e3View commit details
Commits on Apr 20, 2023
-
Fix bug which skipped some utterances if batch_size more than 1
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for f1561d4 - Browse repository at this point
Copy the full SHA f1561d4View commit details
Commits on Apr 21, 2023
-
reduce memory requirements by doing torch.gather on a slice of the lo…
…g probs when they are needed Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 5ebe1e4 - Browse repository at this point
Copy the full SHA 5ebe1e4View commit details
Commits on Apr 22, 2023
-
reduce memory requirements by not saving whole v_matrix
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ccda03e - Browse repository at this point
Copy the full SHA ccda03eView commit details
Commits on May 16, 2023
-
remove any extra spaces in pred_text
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for aad4d04 - Browse repository at this point
Copy the full SHA aad4d04View commit details
Commits on May 22, 2023
-
Merge branch 'main' into nfa_updates
Signed-off-by: Elena Rastorgueva <80532067+erastorgueva-nv@users.noreply.github.com>
Configuration menu - View commit details
-
Copy full SHA for 49031d6 - Browse repository at this point
Copy the full SHA 49031d6View commit details -
[pre-commit.ci] auto fixes from pre-commit.com hooks
for more information, see https://pre-commit.ci
Configuration menu - View commit details
-
Copy full SHA for 033a9fd - Browse repository at this point
Copy the full SHA 033a9fdView commit details -
remove unused list pred_text_all_lines
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for b49ddb7 - Browse repository at this point
Copy the full SHA b49ddb7View commit details
Commits on May 23, 2023
-
support using hybrid Transducer-CTC models for alignment
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for afab69e - Browse repository at this point
Copy the full SHA afab69eView commit details
Commits on Jun 6, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 041b18d - Browse repository at this point
Copy the full SHA 041b18dView commit details -
fix typo - add brackets to torch.cuda.is_available()
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 623369a - Browse repository at this point
Copy the full SHA 623369aView commit details -
make sure token case restoration will work if superscript or subscrip…
…t num is in text Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 2debbc0 - Browse repository at this point
Copy the full SHA 2debbc0View commit details -
Configuration menu - View commit details
-
Copy full SHA for 2879cad - Browse repository at this point
Copy the full SHA 2879cadView commit details -
remove any BOM from input text
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 918abdc - Browse repository at this point
Copy the full SHA 918abdcView commit details -
pick out 1st hypotheses if there is a tuple of them
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 9c91d43 - Browse repository at this point
Copy the full SHA 9c91d43View commit details -
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 846957c - Browse repository at this point
Copy the full SHA 846957cView commit details -
add detail to error message if fail to recover capitalization of tokens
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 0940c70 - Browse repository at this point
Copy the full SHA 0940c70View commit details -
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ed6a4b8 - Browse repository at this point
Copy the full SHA ed6a4b8View commit details -
rename additional_ctm_grouping_separator -> additional_segment_groupi…
…ng_separator Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 412633e - Browse repository at this point
Copy the full SHA 412633eView commit details -
update description of additional_segment_grouping_separator
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ad7df58 - Browse repository at this point
Copy the full SHA ad7df58View commit details -
add simple docstring to get_utt_obj function
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for bd5d274 - Browse repository at this point
Copy the full SHA bd5d274View commit details -
Make docstring for add_t_start_end_to_utt_obj
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for a5f793b - Browse repository at this point
Copy the full SHA a5f793bView commit details -
update docstrings for add_t_start_end_to_utt_obj and get_batch_variables
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 47c72d1 - Browse repository at this point
Copy the full SHA 47c72d1View commit details -
update README and comments in align.py
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for e239375 - Browse repository at this point
Copy the full SHA e239375View commit details -
change 'ground truth' -> 'reference text' in documentation
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for af35b5e - Browse repository at this point
Copy the full SHA af35b5eView commit details
Commits on Jun 7, 2023
-
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ecb3ce2 - Browse repository at this point
Copy the full SHA ecb3ce2View commit details -
add comments to get_utt_obj function
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 93ac8f6 - Browse repository at this point
Copy the full SHA 93ac8f6View commit details -
move constants so they are after imports
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for ce243e0 - Browse repository at this point
Copy the full SHA ce243e0View commit details -
add file description for make_ass_files
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 9827ce4 - Browse repository at this point
Copy the full SHA 9827ce4View commit details -
get rid of Utterance object's S attribute, and correct tests so they …
…pass now Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 5935ada - Browse repository at this point
Copy the full SHA 5935adaView commit details -
Configuration menu - View commit details
-
Copy full SHA for 23380c5 - Browse repository at this point
Copy the full SHA 23380c5View commit details -
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 8ac4a2f - Browse repository at this point
Copy the full SHA 8ac4a2fView commit details -
remove unused variable model from functions saving output files
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 195f306 - Browse repository at this point
Copy the full SHA 195f306View commit details -
remove unused var minimum_timestamp_duration from make_ass_files func…
…tions and return utt_obj Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for b92c25f - Browse repository at this point
Copy the full SHA b92c25fView commit details -
move minimum_timestamp_duration param to CTMFileConfig
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for d3a49e5 - Browse repository at this point
Copy the full SHA d3a49e5View commit details
Commits on Jun 8, 2023
-
Configuration menu - View commit details
-
Copy full SHA for 4f7714c - Browse repository at this point
Copy the full SHA 4f7714cView commit details -
remove unused enumerate and unused import
Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for dab537d - Browse repository at this point
Copy the full SHA dab537dView commit details -
switch reading duration from librosa to soundfile to avoid filename/p…
…ath deprecation message Signed-off-by: Elena Rastorgueva <erastorgueva@nvidia.com>
Configuration menu - View commit details
-
Copy full SHA for 7312497 - Browse repository at this point
Copy the full SHA 7312497View commit details -
Configuration menu - View commit details
-
Copy full SHA for 76cd1b3 - Browse repository at this point
Copy the full SHA 76cd1b3View commit details -
Configuration menu - View commit details
-
Copy full SHA for 6b7959b - Browse repository at this point
Copy the full SHA 6b7959bView commit details -
Configuration menu - View commit details
-
Copy full SHA for 0b6c5f7 - Browse repository at this point
Copy the full SHA 0b6c5f7View commit details -
Configuration menu - View commit details
-
Copy full SHA for 93c0d69 - Browse repository at this point
Copy the full SHA 93c0d69View commit details