MCPcopy Create free account

hub / github.com/JorisCos/LibriMix / functions

Functions49 in github.com/JorisCos/LibriMix

⨍ Functions 49◇ Types & classes 0
↓ 4 callersFunctionaugment_noise
(sound_paths, speed)
scripts/augment_train_noise.py:44
↓ 2 callersFunctionget_list_from_csv
Transform a list in the .csv in an actual python list
scripts/create_librimix_from_metadata.py:262
↓ 2 callersFunctionset_noise_pairs
(pairs, noise_pairs, librispeech_md_file, wham_md_file)
scripts/create_librimix_metadata.py:229
↓ 2 callersFunctionset_utt_pairs
(librispeech_md_file, pair_list, n_src)
scripts/create_librimix_metadata.py:204
↓ 1 callersFunctionadd_noise
(wham_md_file, wham_dir, pair_noise, sources_list, sources_info)
scripts/create_librimix_metadata.py:316
↓ 1 callersFunctionadd_to_metrics_metadata
Add a new line to metrics_df
scripts/create_librimix_from_metadata.py:394
↓ 1 callersFunctionadd_to_mixture_metadata
Add a new line to mixture_df
scripts/create_librimix_from_metadata.py:400
↓ 1 callersFunctioncheck_already_generated
Check if files have already been generated
scripts/create_wham_metadata.py:55
↓ 1 callersFunctioncheck_already_generated
(md_dir, dataset, to_be_ignored, librispeech_md_files)
scripts/create_librimix_metadata.py:115
↓ 1 callersFunctioncheck_already_generated
(md_dir, librispeech_dir)
scripts/create_librispeech_metadata.py:76
↓ 1 callersFunctioncheck_for_cliping
Check the mixture (mode max) for clipping and re normalize if needed.
scripts/create_librimix_metadata.py:392
↓ 1 callersFunctioncompute_gain
Compute the gain between the original and target loudness
scripts/create_librimix_metadata.py:412
↓ 1 callersFunctioncompute_snr_list
Compute the SNR on the mixture mode min
scripts/create_librimix_from_metadata.py:380
↓ 1 callersFunctioncreate_empty_metrics_md
Create the metrics dataframe
scripts/create_librimix_from_metadata.py:194
↓ 1 callersFunctioncreate_empty_mixture_md
Create the mixture dataframe
scripts/create_librimix_from_metadata.py:211
↓ 1 callersFunctioncreate_librimix
Generate sources mixtures and saves them in out_dir
scripts/create_librimix_from_metadata.py:62
↓ 1 callersFunctioncreate_librimix_df
Generate librimix dataframe from a LibriSpeech and wha md file
scripts/create_librimix_metadata.py:135
↓ 1 callersFunctioncreate_librimix_metadata
Generate LibriMix metadata according to LibriSpeech metadata
scripts/create_librimix_metadata.py:58
↓ 1 callersFunctioncreate_librispeech_dataframe
Generate a dataframe that gather infos about the sound files in a LibriSpeech subdirectory
scripts/create_librispeech_metadata.py:94
↓ 1 callersFunctioncreate_librispeech_metadata
Generate metadata corresponding to downloaded data in LibriSpeech
scripts/create_librispeech_metadata.py:28
↓ 1 callersFunctioncreate_speakers_dataframe
Read metadata from the LibriSpeech dataset and collect infos about the speakers
scripts/create_librispeech_metadata.py:51
↓ 1 callersFunctioncreate_wham_noise_dataframe
Generate a dataframe that gather infos about the sound files in a wham_noise subdirectory
scripts/create_wham_metadata.py:71
↓ 1 callersFunctioncreate_wham_noise_metadata
Generate metadata corresponding to downloaded data in wham_noise
scripts/create_wham_metadata.py:27
↓ 1 callersFunctionextend_noise
Concatenate noise using hanning window
scripts/create_librimix_from_metadata.py:273
↓ 1 callersFunctionfit_lengths
Make the sources to match the target length
scripts/create_librimix_from_metadata.py:323
↓ 1 callersFunctionget_row
Get new row for each mixture/info dataframe
scripts/create_librimix_metadata.py:421
↓ 1 callersFunctionloudness_normalize
Normalize sources loudness
scripts/create_librimix_from_metadata.py:304
↓ 1 callersFunctionmain
(args)
scripts/create_librimix_from_metadata.py:37
↓ 1 callersFunctionmain
(args)
scripts/create_wham_metadata.py:19
↓ 1 callersFunctionmain
(args)
scripts/create_librimix_metadata.py:42
↓ 1 callersFunctionmain
(args)
scripts/augment_train_noise.py:15
↓ 1 callersFunctionmain
(args)
scripts/create_librispeech_metadata.py:20
↓ 1 callersFunctionmix
Do the mixing
scripts/create_librimix_from_metadata.py:362
↓ 1 callersFunctionmix
Do the mixture for min mode and max mode
scripts/create_librimix_metadata.py:383
↓ 1 callersFunctionprocess_metadata_file
Process a metadata generation file to create sources and mixtures
scripts/create_librimix_from_metadata.py:75
↓ 1 callersFunctionprocess_utterances
(md_file, librispeech_dir, wham_dir, freq, mode, subdirs, dir_path, subset_metadata_pat
scripts/create_librimix_from_metadata.py:118
↓ 1 callersFunctionread_sources
Get sources and info to mix the sources
scripts/create_librimix_from_metadata.py:230
↓ 1 callersFunctionread_sources
(metadata_file, pair, n_src, librispeech_dir)
scripts/create_librimix_metadata.py:285
↓ 1 callersFunctionremove_duplicates
(utt_pairs, noise_pairs)
scripts/create_librimix_metadata.py:272
↓ 1 callersFunctionresample_list
Resample the source list to the desired frequency
scripts/create_librimix_from_metadata.py:313
↓ 1 callersFunctionset_loudness
Compute original loudness and normalise them randomly
scripts/create_librimix_metadata.py:345
↓ 1 callersFunctionset_pairs
set pairs of sources to make the mixture
scripts/create_librimix_metadata.py:181
↓ 1 callersFunctionsnr_xy
(x, y)
scripts/create_librimix_from_metadata.py:390
↓ 1 callersFunctiontransform_sources
Transform libriSpeech sources to librimix
scripts/create_librimix_from_metadata.py:293
↓ 1 callersFunctionwrite_mix
(mix_id, mixture, dir_path, subdir, freq)
scripts/create_librimix_from_metadata.py:371
↓ 1 callersFunctionwrite_noise
(mix_id, transformed_sources, dir_path, freq)
scripts/create_librimix_from_metadata.py:352
↓ 1 callersFunctionwrite_sources
(mix_id, transformed_sources, subdirs, dir_path, freq, n_src)
scripts/create_librimix_from_metadata.py:340
Functionapply_fx
(sound_path, speed)
scripts/augment_train_noise.py:53
Functionprocess_utterance
(n_src, librispeech_dir, wham_dir, freq, mode, subdirs, dir_path, row)
scripts/create_librimix_from_metadata.py:155