Upload 4 files
Browse files
.gitattributes
CHANGED
|
@@ -51,3 +51,4 @@ eval_results/singing_librispeech/examples/ex_69/s1.wav filter=lfs diff=lfs merge
|
|
| 51 |
vocal[[:space:]]231/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
| 52 |
vocals[[:space:]]238/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
| 53 |
singing_librispeech_ft_iSRNet/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 51 |
vocal[[:space:]]231/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
| 52 |
vocals[[:space:]]238/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
| 53 |
singing_librispeech_ft_iSRNet/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
multi_singing_librispeech/vocals.chkpnt filter=lfs diff=lfs merge=lfs -text
|
multi_singing_librispeech/loss_graph_vocals.png
ADDED
|
multi_singing_librispeech/vocals.chkpnt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37b80a59b1226ebca7dfc8ffa6db410ba683e328b5ba422f9bee7051c85614cc
|
| 3 |
+
size 396710313
|
multi_singing_librispeech/vocals.json
ADDED
|
@@ -0,0 +1,642 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"args": {
|
| 3 |
+
"above_freq": 300.0,
|
| 4 |
+
"architecture": "conv_tasnet_stft",
|
| 5 |
+
"batch_size": 58,
|
| 6 |
+
"beta1": 0.5,
|
| 7 |
+
"beta2": 0.9,
|
| 8 |
+
"bn_chan": 256,
|
| 9 |
+
"continual_train": false,
|
| 10 |
+
"dataset": "multi_singing_librispeech",
|
| 11 |
+
"db_normalize": false,
|
| 12 |
+
"ema": true,
|
| 13 |
+
"encoder_activation": null,
|
| 14 |
+
"entity": "carson2050",
|
| 15 |
+
"epochs": 200,
|
| 16 |
+
"eps": 1e-08,
|
| 17 |
+
"exp_name": "multi_singing_librispeech",
|
| 18 |
+
"ff_activation": "relu",
|
| 19 |
+
"gpu": 0,
|
| 20 |
+
"gradient_clip": null,
|
| 21 |
+
"hid_chan": 1024,
|
| 22 |
+
"load_ema_online_model": false,
|
| 23 |
+
"lr": 0.0002,
|
| 24 |
+
"lr_decay_gamma": 0.5,
|
| 25 |
+
"lr_decay_patience": 20,
|
| 26 |
+
"lr_scheduler": "step_lr",
|
| 27 |
+
"mask_act": "linear",
|
| 28 |
+
"max_n_src": 4,
|
| 29 |
+
"min_n_src": 2,
|
| 30 |
+
"mixed_precision": false,
|
| 31 |
+
"mixture_consistency": "mixture_consistency",
|
| 32 |
+
"multi_spec_loss_log_scale": false,
|
| 33 |
+
"n_blocks": 6,
|
| 34 |
+
"n_filter": 512,
|
| 35 |
+
"n_kernel": 512,
|
| 36 |
+
"n_nodes": 1,
|
| 37 |
+
"n_repeats": 4,
|
| 38 |
+
"n_src": 2,
|
| 39 |
+
"nb_workers": 8,
|
| 40 |
+
"nfft": 2048,
|
| 41 |
+
"ngpus_per_node": 1,
|
| 42 |
+
"nhop": 512,
|
| 43 |
+
"no_cuda": false,
|
| 44 |
+
"no_mask": false,
|
| 45 |
+
"no_mask_residual": false,
|
| 46 |
+
"optimizer": "adam",
|
| 47 |
+
"output": "results/singing_sep/checkpoint/multi_singing_librispeech",
|
| 48 |
+
"output_directory": "results/singing_sep",
|
| 49 |
+
"part_of_data": null,
|
| 50 |
+
"patience": 50,
|
| 51 |
+
"pitch_formant_augment_prob": 0.4,
|
| 52 |
+
"port": null,
|
| 53 |
+
"project": "MedleyVox_home_pt2",
|
| 54 |
+
"quiet": false,
|
| 55 |
+
"rank": 0,
|
| 56 |
+
"reduced_training_data_ratio": 1.0,
|
| 57 |
+
"resume": "results/singing_sep/checkpoint/multi_singing_librispeech",
|
| 58 |
+
"same_singer_dict_path": [
|
| 59 |
+
[
|
| 60 |
+
"../data/24k/OpenSinger",
|
| 61 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_OpenSinger.json",
|
| 62 |
+
"OpenSinger"
|
| 63 |
+
],
|
| 64 |
+
[
|
| 65 |
+
"../data/24k/k_multisinger",
|
| 66 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_k_multisinger.json",
|
| 67 |
+
"k_multisinger"
|
| 68 |
+
],
|
| 69 |
+
[
|
| 70 |
+
"../data/24k/CSD",
|
| 71 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_CSD.json",
|
| 72 |
+
"CSD"
|
| 73 |
+
],
|
| 74 |
+
[
|
| 75 |
+
"../data/24k/jsut-song_ver1",
|
| 76 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_jsut-song_ver1.json",
|
| 77 |
+
"jsut-song_ver1"
|
| 78 |
+
],
|
| 79 |
+
[
|
| 80 |
+
"../data/24k/jvs_music_ver1",
|
| 81 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_jvs_music_ver1.json",
|
| 82 |
+
"jvs_music_ver1"
|
| 83 |
+
],
|
| 84 |
+
[
|
| 85 |
+
"../data/24k/k_multitimbre",
|
| 86 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_k_multitimbre.json",
|
| 87 |
+
"k_multitimbre"
|
| 88 |
+
],
|
| 89 |
+
[
|
| 90 |
+
"../data/24k/kiritan_revised",
|
| 91 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_kiritan.json",
|
| 92 |
+
"kiritan"
|
| 93 |
+
],
|
| 94 |
+
[
|
| 95 |
+
"../data/24k/musdb_a_train",
|
| 96 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_musdb_a_train.json",
|
| 97 |
+
"musdb_a_train"
|
| 98 |
+
],
|
| 99 |
+
[
|
| 100 |
+
"../data/24k/NUS",
|
| 101 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_NUS.json",
|
| 102 |
+
"NUS"
|
| 103 |
+
],
|
| 104 |
+
[
|
| 105 |
+
"../data/24k/VocalSet",
|
| 106 |
+
"./svs/preprocess/make_same_singer_dict/same_singer_VocalSet.json",
|
| 107 |
+
"VocalSet"
|
| 108 |
+
]
|
| 109 |
+
],
|
| 110 |
+
"same_singer_ratio": 0.2,
|
| 111 |
+
"same_song_dict_path": [
|
| 112 |
+
[
|
| 113 |
+
"../data/24k/k_multisinger",
|
| 114 |
+
"./svs/preprocess/make_same_song_dict/same_song_k_multisinger.json",
|
| 115 |
+
"k_multisinger"
|
| 116 |
+
]
|
| 117 |
+
],
|
| 118 |
+
"same_song_ratio": 0.2,
|
| 119 |
+
"same_speaker_dict_path": [
|
| 120 |
+
[
|
| 121 |
+
"../data/24k/LibriSpeech_train-clean-100",
|
| 122 |
+
"./svs/preprocess/make_same_speaker_dict/same_singer_LibriSpeech_train-clean-100.json",
|
| 123 |
+
"LibriSpeech_train-clean-100"
|
| 124 |
+
],
|
| 125 |
+
[
|
| 126 |
+
"../data/24k/LibriSpeech_train-clean-360",
|
| 127 |
+
"./svs/preprocess/make_same_speaker_dict/same_singer_LibriSpeech_train-clean-360.json",
|
| 128 |
+
"LibriSpeech_train-clean-360"
|
| 129 |
+
]
|
| 130 |
+
],
|
| 131 |
+
"same_speaker_ratio": 0.15,
|
| 132 |
+
"sample_rate": 24000,
|
| 133 |
+
"seed": 777,
|
| 134 |
+
"seq_dur": 3.0,
|
| 135 |
+
"sing_sing_ratio": 0.15,
|
| 136 |
+
"sing_speech_ratio": 0.15,
|
| 137 |
+
"skip_chan": 256,
|
| 138 |
+
"song_length_dict_path": "./svs/preprocess/song_length_dict_24k.json",
|
| 139 |
+
"speech_train_root": [
|
| 140 |
+
"../data/24k/LibriSpeech_train-clean-360",
|
| 141 |
+
"../data/24k/LibriSpeech_train-clean-100"
|
| 142 |
+
],
|
| 143 |
+
"sr_input_res": false,
|
| 144 |
+
"sr_out_mix_consistency": false,
|
| 145 |
+
"srnet": "orig",
|
| 146 |
+
"start_from_best": false,
|
| 147 |
+
"sweep": false,
|
| 148 |
+
"target": "vocals",
|
| 149 |
+
"train_loss_func": [
|
| 150 |
+
"pit_snr",
|
| 151 |
+
"multi_spectral_l1"
|
| 152 |
+
],
|
| 153 |
+
"train_root": [
|
| 154 |
+
"../data/24k/CSD",
|
| 155 |
+
"../data/24k/NUS",
|
| 156 |
+
"../data/24k/TONAS",
|
| 157 |
+
"../data/24k/VocalSet",
|
| 158 |
+
"../data/24k/jsut-song_ver1",
|
| 159 |
+
"../data/24k/jvs_music_ver1",
|
| 160 |
+
"../data/24k/kiritan_revised",
|
| 161 |
+
"../data/24k/vocadito",
|
| 162 |
+
"../data/24k/musdb_a_train",
|
| 163 |
+
"../data/24k/OpenSinger",
|
| 164 |
+
"../data/24k/medleyDB_v1_in_musdb",
|
| 165 |
+
"../data/24k/k_multisinger",
|
| 166 |
+
"../data/24k/k_multitimbre"
|
| 167 |
+
],
|
| 168 |
+
"unison_prob": 0.3,
|
| 169 |
+
"use_wandb": true,
|
| 170 |
+
"valid_loss_func": [
|
| 171 |
+
"pit_si_sdr"
|
| 172 |
+
],
|
| 173 |
+
"valid_regions_dict_path": "./svs/preprocess/valid_regions_dict_singing_singing.json",
|
| 174 |
+
"valid_root": [
|
| 175 |
+
[
|
| 176 |
+
"../data/24k/musdb_a_test",
|
| 177 |
+
"../data/24k/musdb_a_test",
|
| 178 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_singing_singing.json",
|
| 179 |
+
"sing_sing_diff"
|
| 180 |
+
],
|
| 181 |
+
[
|
| 182 |
+
"../data/24k/musdb_a_test",
|
| 183 |
+
"../data/24k/musdb_a_test",
|
| 184 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_singing_unison.json",
|
| 185 |
+
"sing_sing_unison"
|
| 186 |
+
],
|
| 187 |
+
[
|
| 188 |
+
"../data/24k/musdb_a_test",
|
| 189 |
+
"../data/24k/musdb_a_test",
|
| 190 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_singing_singing_same_singer.json",
|
| 191 |
+
"sing_sing_same_singer"
|
| 192 |
+
],
|
| 193 |
+
[
|
| 194 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 195 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 196 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_speech_speech.json",
|
| 197 |
+
"speech_speech_diff"
|
| 198 |
+
],
|
| 199 |
+
[
|
| 200 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 201 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 202 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_speech_unison.json",
|
| 203 |
+
"speech_speech_unison"
|
| 204 |
+
],
|
| 205 |
+
[
|
| 206 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 207 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 208 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_speech_speech_same_speaker.json",
|
| 209 |
+
"speech_speech_same_speaker"
|
| 210 |
+
],
|
| 211 |
+
[
|
| 212 |
+
"../data/24k/musdb_a_test",
|
| 213 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 214 |
+
"./svs/preprocess/make_validation_dict/for_2_srcs/valid_regions_dict_singing_speech.json",
|
| 215 |
+
"singing_speech"
|
| 216 |
+
]
|
| 217 |
+
],
|
| 218 |
+
"valid_root_orpit": [
|
| 219 |
+
[
|
| 220 |
+
"../data/24k/musdb_a_test",
|
| 221 |
+
"../data/24k/musdb_a_test",
|
| 222 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_singing_singing_n_srcs.json",
|
| 223 |
+
"sing_sing_diff"
|
| 224 |
+
],
|
| 225 |
+
[
|
| 226 |
+
"../data/24k/musdb_a_test",
|
| 227 |
+
"../data/24k/musdb_a_test",
|
| 228 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_singing_unison_n_srcs.json",
|
| 229 |
+
"sing_sing_unison"
|
| 230 |
+
],
|
| 231 |
+
[
|
| 232 |
+
"../data/24k/musdb_a_test",
|
| 233 |
+
"../data/24k/musdb_a_test",
|
| 234 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_singing_singing_same_singer_n_srcs.json",
|
| 235 |
+
"sing_sing_same_singer"
|
| 236 |
+
],
|
| 237 |
+
[
|
| 238 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 239 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 240 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_speech_speech_n_srcs.json",
|
| 241 |
+
"speech_speech_diff"
|
| 242 |
+
],
|
| 243 |
+
[
|
| 244 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 245 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 246 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_speech_unison_n_srcs.json",
|
| 247 |
+
"speech_speech_unison"
|
| 248 |
+
],
|
| 249 |
+
[
|
| 250 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 251 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 252 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_speech_speech_same_speaker_n_srcs.json",
|
| 253 |
+
"speech_speech_same_speaker"
|
| 254 |
+
],
|
| 255 |
+
[
|
| 256 |
+
"../data/24k/musdb_a_test",
|
| 257 |
+
"../data/24k/LibriSpeech_dev-clean",
|
| 258 |
+
"./svs/preprocess/make_validation_dict/for_n_srcs/valid_regions_dict_singing_speech_n_srcs.json",
|
| 259 |
+
"singing_speech"
|
| 260 |
+
]
|
| 261 |
+
],
|
| 262 |
+
"weight_decay": 1e-06,
|
| 263 |
+
"world_size": 1
|
| 264 |
+
},
|
| 265 |
+
"best_epoch": 92,
|
| 266 |
+
"best_loss": -7.675145898546491,
|
| 267 |
+
"epochs_trained": 93,
|
| 268 |
+
"num_bad_epochs": 1,
|
| 269 |
+
"train_loss_history": [
|
| 270 |
+
-1.4921716451644897,
|
| 271 |
+
-2.8664329051971436,
|
| 272 |
+
-3.3393054008483887,
|
| 273 |
+
-3.638092517852783,
|
| 274 |
+
-3.879303216934204,
|
| 275 |
+
-4.088993072509766,
|
| 276 |
+
-4.227543830871582,
|
| 277 |
+
-4.412248134613037,
|
| 278 |
+
-4.58026123046875,
|
| 279 |
+
-4.71837043762207,
|
| 280 |
+
-4.800468444824219,
|
| 281 |
+
-4.882855415344238,
|
| 282 |
+
-5.011181831359863,
|
| 283 |
+
-5.128243923187256,
|
| 284 |
+
-5.150334358215332,
|
| 285 |
+
-5.240769386291504,
|
| 286 |
+
-5.357062816619873,
|
| 287 |
+
-5.35420560836792,
|
| 288 |
+
-5.427922248840332,
|
| 289 |
+
-5.536999225616455,
|
| 290 |
+
-5.6000895500183105,
|
| 291 |
+
-5.664849758148193,
|
| 292 |
+
-5.704154968261719,
|
| 293 |
+
-5.791101455688477,
|
| 294 |
+
-5.794349670410156,
|
| 295 |
+
-5.784161567687988,
|
| 296 |
+
-5.824007511138916,
|
| 297 |
+
-5.931461811065674,
|
| 298 |
+
-5.981809139251709,
|
| 299 |
+
-6.045787334442139,
|
| 300 |
+
-6.045494079589844,
|
| 301 |
+
-6.075621128082275,
|
| 302 |
+
-6.086508750915527,
|
| 303 |
+
-6.123781681060791,
|
| 304 |
+
-6.192169666290283,
|
| 305 |
+
-6.248963832855225,
|
| 306 |
+
-6.183308124542236,
|
| 307 |
+
-6.25191593170166,
|
| 308 |
+
-6.301548004150391,
|
| 309 |
+
-6.259702682495117,
|
| 310 |
+
-6.338959217071533,
|
| 311 |
+
-6.372439861297607,
|
| 312 |
+
-6.436537742614746,
|
| 313 |
+
-6.462899684906006,
|
| 314 |
+
-6.449411392211914,
|
| 315 |
+
-6.489621639251709,
|
| 316 |
+
-6.461447238922119,
|
| 317 |
+
-6.558005332946777,
|
| 318 |
+
-6.603482723236084,
|
| 319 |
+
-6.555445194244385,
|
| 320 |
+
-6.579801082611084,
|
| 321 |
+
-6.639071464538574,
|
| 322 |
+
-6.648660182952881,
|
| 323 |
+
-6.6866631507873535,
|
| 324 |
+
-6.767474174499512,
|
| 325 |
+
-6.750443935394287,
|
| 326 |
+
-6.7306742668151855,
|
| 327 |
+
-6.7853617668151855,
|
| 328 |
+
-6.818509101867676,
|
| 329 |
+
-6.761360168457031,
|
| 330 |
+
-6.795668601989746,
|
| 331 |
+
-6.82188606262207,
|
| 332 |
+
-6.795504093170166,
|
| 333 |
+
-6.914917469024658,
|
| 334 |
+
-6.921656131744385,
|
| 335 |
+
-6.950718402862549,
|
| 336 |
+
-6.966548919677734,
|
| 337 |
+
-6.965545654296875,
|
| 338 |
+
-6.964168548583984,
|
| 339 |
+
-6.888548374176025,
|
| 340 |
+
-6.932443141937256,
|
| 341 |
+
-6.930734634399414,
|
| 342 |
+
-6.949597358703613,
|
| 343 |
+
-6.947843551635742,
|
| 344 |
+
-6.959360599517822,
|
| 345 |
+
-6.974522590637207,
|
| 346 |
+
-7.005373954772949,
|
| 347 |
+
-7.039368629455566,
|
| 348 |
+
-7.008696556091309,
|
| 349 |
+
-7.064368724822998,
|
| 350 |
+
-7.038439750671387,
|
| 351 |
+
-7.046519756317139,
|
| 352 |
+
-7.052777290344238,
|
| 353 |
+
-7.06027889251709,
|
| 354 |
+
-7.048835277557373,
|
| 355 |
+
-7.095891952514648,
|
| 356 |
+
-7.080573558807373,
|
| 357 |
+
-7.120383262634277,
|
| 358 |
+
-7.1035075187683105,
|
| 359 |
+
-7.147456645965576,
|
| 360 |
+
-7.133329391479492,
|
| 361 |
+
-7.134939670562744,
|
| 362 |
+
-7.155049800872803
|
| 363 |
+
],
|
| 364 |
+
"train_time_history": [
|
| 365 |
+
4810.419310808182,
|
| 366 |
+
4810.429551362991,
|
| 367 |
+
4780.074353456497,
|
| 368 |
+
4780.08434343338,
|
| 369 |
+
4793.850719213486,
|
| 370 |
+
4793.851686954498,
|
| 371 |
+
4799.262031078339,
|
| 372 |
+
4799.2719786167145,
|
| 373 |
+
4776.265509605408,
|
| 374 |
+
4776.275769710541,
|
| 375 |
+
4800.915772199631,
|
| 376 |
+
4800.925550937653,
|
| 377 |
+
4782.19565987587,
|
| 378 |
+
4870.2729279994965,
|
| 379 |
+
4864.202353715897,
|
| 380 |
+
5526.39341044426,
|
| 381 |
+
5526.402764797211,
|
| 382 |
+
5210.057184457779,
|
| 383 |
+
5210.0663821697235,
|
| 384 |
+
5192.114199876785,
|
| 385 |
+
5192.115474700928,
|
| 386 |
+
5119.568732976913,
|
| 387 |
+
5119.579450130463,
|
| 388 |
+
4854.391019105911,
|
| 389 |
+
4854.4009165763855,
|
| 390 |
+
4825.207883834839,
|
| 391 |
+
4825.218036174774,
|
| 392 |
+
4839.370161294937,
|
| 393 |
+
4839.3797080516815,
|
| 394 |
+
4829.168277978897,
|
| 395 |
+
4829.178178310394,
|
| 396 |
+
4831.754481077194,
|
| 397 |
+
4831.764403104782,
|
| 398 |
+
4840.167069673538,
|
| 399 |
+
4840.1764142513275,
|
| 400 |
+
4839.306309938431,
|
| 401 |
+
4839.315984725952,
|
| 402 |
+
4835.479310274124,
|
| 403 |
+
4835.489530324936,
|
| 404 |
+
4991.815203428268,
|
| 405 |
+
4826.761980772018,
|
| 406 |
+
4826.772101163864,
|
| 407 |
+
4819.3496108055115,
|
| 408 |
+
4819.358952999115,
|
| 409 |
+
4820.984974384308,
|
| 410 |
+
4820.995014190674,
|
| 411 |
+
4820.539752483368,
|
| 412 |
+
4820.548979997635,
|
| 413 |
+
4814.873534917831,
|
| 414 |
+
4814.876268863678,
|
| 415 |
+
4812.354250907898,
|
| 416 |
+
4812.363839626312,
|
| 417 |
+
4822.391925573349,
|
| 418 |
+
4822.40118765831,
|
| 419 |
+
4809.686738491058,
|
| 420 |
+
4809.697638034821,
|
| 421 |
+
4832.5055372715,
|
| 422 |
+
4832.515355587006,
|
| 423 |
+
4831.67563867569,
|
| 424 |
+
4831.685403108597,
|
| 425 |
+
4824.845934391022,
|
| 426 |
+
4824.85514998436,
|
| 427 |
+
4835.57625246048,
|
| 428 |
+
4835.587289094925,
|
| 429 |
+
4817.744952201843,
|
| 430 |
+
4817.7542552948,
|
| 431 |
+
4807.804133653641,
|
| 432 |
+
4807.814810037613,
|
| 433 |
+
4818.521605968475,
|
| 434 |
+
4818.532015800476,
|
| 435 |
+
4981.354954957962,
|
| 436 |
+
4981.368631839752,
|
| 437 |
+
4875.586889028549,
|
| 438 |
+
4875.597553014755,
|
| 439 |
+
4801.111567258835,
|
| 440 |
+
4801.1219182014465,
|
| 441 |
+
4799.074081897736,
|
| 442 |
+
4799.08514547348,
|
| 443 |
+
4809.401276350021,
|
| 444 |
+
4809.41465306282,
|
| 445 |
+
4809.043102502823,
|
| 446 |
+
4809.04475402832,
|
| 447 |
+
4818.2070748806,
|
| 448 |
+
4818.209503889084,
|
| 449 |
+
4796.3679666519165,
|
| 450 |
+
4796.377726793289,
|
| 451 |
+
4794.153427362442,
|
| 452 |
+
4794.155965805054,
|
| 453 |
+
4804.1572597026825,
|
| 454 |
+
4804.168130159378,
|
| 455 |
+
4797.392125368118,
|
| 456 |
+
4797.401923418045,
|
| 457 |
+
4797.116873264313,
|
| 458 |
+
4797.12747836113,
|
| 459 |
+
4799.205674409866,
|
| 460 |
+
4799.215870857239,
|
| 461 |
+
4969.960748910904,
|
| 462 |
+
4969.971879482269,
|
| 463 |
+
5270.599810838699,
|
| 464 |
+
5270.6101796627045,
|
| 465 |
+
4881.989181518555,
|
| 466 |
+
4882.000226974487,
|
| 467 |
+
4867.6136746406555,
|
| 468 |
+
4867.624637126923,
|
| 469 |
+
5128.904933452606,
|
| 470 |
+
5128.915862798691,
|
| 471 |
+
4879.79870891571,
|
| 472 |
+
4879.80947971344,
|
| 473 |
+
4969.744366407394,
|
| 474 |
+
4969.754128456116,
|
| 475 |
+
4907.097052812576,
|
| 476 |
+
4907.107843637466,
|
| 477 |
+
4812.9132516384125,
|
| 478 |
+
4812.9242560863495,
|
| 479 |
+
4815.909214496613,
|
| 480 |
+
4815.920344591141,
|
| 481 |
+
4806.699935913086,
|
| 482 |
+
4806.70260477066,
|
| 483 |
+
4831.170897245407,
|
| 484 |
+
4831.180289506912,
|
| 485 |
+
4839.252681255341,
|
| 486 |
+
4839.262135982513,
|
| 487 |
+
4833.886634111404,
|
| 488 |
+
4833.8898758888245,
|
| 489 |
+
4830.524186134338,
|
| 490 |
+
4830.53564286232,
|
| 491 |
+
4824.74093079567,
|
| 492 |
+
4824.747734546661,
|
| 493 |
+
4818.754670858383,
|
| 494 |
+
4818.764072179794,
|
| 495 |
+
4816.966838121414,
|
| 496 |
+
4816.977759599686,
|
| 497 |
+
4831.582427740097,
|
| 498 |
+
4831.592094898224,
|
| 499 |
+
4804.51261639595,
|
| 500 |
+
4993.473606586456,
|
| 501 |
+
4815.601177692413,
|
| 502 |
+
4815.610737085342,
|
| 503 |
+
4789.788247346878,
|
| 504 |
+
4865.854624032974,
|
| 505 |
+
4865.864605426788,
|
| 506 |
+
5030.177618980408,
|
| 507 |
+
5030.188777208328,
|
| 508 |
+
4769.904754638672,
|
| 509 |
+
4895.086503267288,
|
| 510 |
+
4895.0979063510895,
|
| 511 |
+
4869.957269668579,
|
| 512 |
+
4869.958615779877,
|
| 513 |
+
4962.7930123806,
|
| 514 |
+
4962.803097486496,
|
| 515 |
+
5163.898764133453,
|
| 516 |
+
5163.90244436264,
|
| 517 |
+
4803.290739297867,
|
| 518 |
+
4803.3002672195435,
|
| 519 |
+
4819.002298593521,
|
| 520 |
+
4819.014036178589,
|
| 521 |
+
4812.0083973407745,
|
| 522 |
+
5130.011174440384,
|
| 523 |
+
5130.020927429199,
|
| 524 |
+
5238.748838424683,
|
| 525 |
+
5162.933927536011,
|
| 526 |
+
5162.94544506073,
|
| 527 |
+
5014.217702865601,
|
| 528 |
+
5014.227581739426,
|
| 529 |
+
5119.955267906189,
|
| 530 |
+
5119.966482877731,
|
| 531 |
+
4877.71505856514,
|
| 532 |
+
4947.4076771736145,
|
| 533 |
+
4947.418792486191,
|
| 534 |
+
4980.132425069809,
|
| 535 |
+
4980.143876552582,
|
| 536 |
+
5166.483239412308,
|
| 537 |
+
5166.49423623085,
|
| 538 |
+
4906.088274717331,
|
| 539 |
+
4906.0993638038635,
|
| 540 |
+
4880.329564332962,
|
| 541 |
+
4880.339328289032,
|
| 542 |
+
4873.104112148285,
|
| 543 |
+
4873.119816303253,
|
| 544 |
+
4885.143585205078,
|
| 545 |
+
5015.694309234619
|
| 546 |
+
],
|
| 547 |
+
"valid_loss_history": [
|
| 548 |
+
-3.0233164174216136,
|
| 549 |
+
-4.020770004817417,
|
| 550 |
+
-4.493505137307303,
|
| 551 |
+
-4.805826323372977,
|
| 552 |
+
-5.009723663330078,
|
| 553 |
+
-5.3551515851702005,
|
| 554 |
+
-5.507791314806257,
|
| 555 |
+
-5.709285395486014,
|
| 556 |
+
-5.82812111718314,
|
| 557 |
+
-5.88963999067034,
|
| 558 |
+
-5.970332486288888,
|
| 559 |
+
-5.988547257014683,
|
| 560 |
+
-6.113276481628418,
|
| 561 |
+
-6.132954188755581,
|
| 562 |
+
-6.210943358285086,
|
| 563 |
+
-6.279647214072091,
|
| 564 |
+
-6.3300862312316895,
|
| 565 |
+
-6.36109277180263,
|
| 566 |
+
-6.427222183772495,
|
| 567 |
+
-6.453009741646903,
|
| 568 |
+
-6.489914894104004,
|
| 569 |
+
-6.48867974962507,
|
| 570 |
+
-6.536007336207798,
|
| 571 |
+
-6.532879625047956,
|
| 572 |
+
-6.572577135903495,
|
| 573 |
+
-6.5566478456769675,
|
| 574 |
+
-6.660695620945522,
|
| 575 |
+
-6.70451055254255,
|
| 576 |
+
-6.65756470816476,
|
| 577 |
+
-6.701659406934466,
|
| 578 |
+
-6.7815567425319125,
|
| 579 |
+
-6.804818085261753,
|
| 580 |
+
-6.783689567020962,
|
| 581 |
+
-6.844764641353062,
|
| 582 |
+
-6.868685790470669,
|
| 583 |
+
-6.888231481824603,
|
| 584 |
+
-6.942419528961182,
|
| 585 |
+
-6.951289176940918,
|
| 586 |
+
-6.975076675415039,
|
| 587 |
+
-6.991657052721296,
|
| 588 |
+
-7.000387941087995,
|
| 589 |
+
-7.082024574279785,
|
| 590 |
+
-7.087371553693499,
|
| 591 |
+
-7.114969117300851,
|
| 592 |
+
-7.163520812988281,
|
| 593 |
+
-7.1951784406389505,
|
| 594 |
+
-7.216815676007952,
|
| 595 |
+
-7.201807635171073,
|
| 596 |
+
-7.183896745954241,
|
| 597 |
+
-7.227273804800851,
|
| 598 |
+
-7.232961927141462,
|
| 599 |
+
-7.2955668313162665,
|
| 600 |
+
-7.2690509387425015,
|
| 601 |
+
-7.273542472294399,
|
| 602 |
+
-7.281754766191755,
|
| 603 |
+
-7.311358247484479,
|
| 604 |
+
-7.287418706076486,
|
| 605 |
+
-7.261871746608189,
|
| 606 |
+
-7.2840664727347235,
|
| 607 |
+
-7.316314697265625,
|
| 608 |
+
-7.376913070678711,
|
| 609 |
+
-7.367326668330601,
|
| 610 |
+
-7.438824789864676,
|
| 611 |
+
-7.427623748779297,
|
| 612 |
+
-7.45092739377703,
|
| 613 |
+
-7.4810590744018555,
|
| 614 |
+
-7.43196405683245,
|
| 615 |
+
-7.413298266274588,
|
| 616 |
+
-7.448171275002616,
|
| 617 |
+
-7.470413276127407,
|
| 618 |
+
-7.478131294250488,
|
| 619 |
+
-7.494483879634312,
|
| 620 |
+
-7.459411212376186,
|
| 621 |
+
-7.515866688319615,
|
| 622 |
+
-7.571803229195731,
|
| 623 |
+
-7.573634147644043,
|
| 624 |
+
-7.562024729592459,
|
| 625 |
+
-7.550929818834577,
|
| 626 |
+
-7.53609037399292,
|
| 627 |
+
-7.563671180180141,
|
| 628 |
+
-7.578108038221087,
|
| 629 |
+
-7.531997203826904,
|
| 630 |
+
-7.600094999585833,
|
| 631 |
+
-7.639314787728446,
|
| 632 |
+
-7.65882362638201,
|
| 633 |
+
-7.624989918300083,
|
| 634 |
+
-7.5906588690621515,
|
| 635 |
+
-7.606484276907785,
|
| 636 |
+
-7.59099394934518,
|
| 637 |
+
-7.611005442483084,
|
| 638 |
+
-7.649059908730643,
|
| 639 |
+
-7.675145898546491,
|
| 640 |
+
-7.672644138336182
|
| 641 |
+
]
|
| 642 |
+
}
|
multi_singing_librispeech/vocals.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5c8ff43108ee58ffc9555359fa05727534d4448d95d58b563332026487de43c8
|
| 3 |
+
size 232997719
|