Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- .gitattributes +19 -0
- Dereverb-Echo-MelBand-Sucial/scripts/create_reverb_delay.py +76 -0
- Dereverb-Echo-MelBand-Sucial/scripts/create_reverb_delay_multi_threads.py +71 -0
- Dereverb-Echo-MelBand-Sucial/scripts/model_fusion.py +13 -0
- Dereverb-MelBand-Anvuew/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.ckpt +3 -0
- Dereverb-MelBand-Anvuew/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.yaml +77 -0
- Dereverb-MelBand-Anvuew/archive only/8_512_12/deverb_mel_band_roformer_8_512dim_12depth.yaml +77 -0
- HuggingFace_Repos/anvuew/BS-RoFormer/.gitattributes +35 -0
- HuggingFace_Repos/anvuew/BS-RoFormer/README.md +5 -0
- HuggingFace_Repos/anvuew/BS-RoFormer/bs_roformer_anvuew_sdr_12.45.ckpt +3 -0
- HuggingFace_Repos/anvuew/BS-RoFormer/config.yaml +129 -0
- HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/.gitattributes +35 -0
- HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/README.md +15 -0
- HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.yaml +77 -0
- HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/archive only/8_512_12/deverb_mel_band_roformer_8_512dim_12depth.yaml +77 -0
- HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/dereverb_mel_band_roformer_anvuew.yaml +76 -0
- HuggingFace_Repos/anvuew/karaoke_bs_roformer/.gitattributes +35 -0
- HuggingFace_Repos/anvuew/karaoke_bs_roformer/README.md +5 -0
- HuggingFace_Repos/anvuew/karaoke_bs_roformer/karaoke_bs_roformer_anvuew.ckpt +3 -0
- HuggingFace_Repos/anvuew/karaoke_bs_roformer/karaoke_bs_roformer_anvuew.yaml +131 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/.~lock.all_metrics.csv# +1 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/all_metrics.csv +117 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/mixture.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/mixture.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/mixture.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1_estimate.wav +3 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_34/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_40/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_61/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_65/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_69/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_72/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_79/metrics.json +13 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech/final_metrics.json +12 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/all_metrics.csv +117 -0
- MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_16/metrics.json +13 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,22 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_92/s0.wav filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_92/s0_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_92/s1.wav filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0.wav filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_92/s1_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/mixture.wav filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1.wav filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 46 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0.wav filter=lfs diff=lfs merge=lfs -text
|
| 47 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/mixture.wav filter=lfs diff=lfs merge=lfs -text
|
| 48 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1.wav filter=lfs diff=lfs merge=lfs -text
|
| 49 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 50 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/mixture.wav filter=lfs diff=lfs merge=lfs -text
|
| 51 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1.wav filter=lfs diff=lfs merge=lfs -text
|
| 52 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
| 53 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0.wav filter=lfs diff=lfs merge=lfs -text
|
| 54 |
+
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1_estimate.wav filter=lfs diff=lfs merge=lfs -text
|
Dereverb-Echo-MelBand-Sucial/scripts/create_reverb_delay.py
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import argparse
|
| 3 |
+
import librosa
|
| 4 |
+
import numpy as np
|
| 5 |
+
import soundfile as sf
|
| 6 |
+
from pedalboard import Pedalboard, Reverb, Delay, HighpassFilter, LowpassFilter
|
| 7 |
+
from random import uniform
|
| 8 |
+
from tqdm import tqdm
|
| 9 |
+
|
| 10 |
+
|
| 11 |
+
def random_effect(audio, sr):
|
| 12 |
+
reverb = Pedalboard([
|
| 13 |
+
Delay(
|
| 14 |
+
delay_seconds=uniform(0.001, 0.100),
|
| 15 |
+
feedback=0.0,
|
| 16 |
+
mix=1.0
|
| 17 |
+
),
|
| 18 |
+
Reverb(
|
| 19 |
+
room_size=uniform(0.1, 0.8),
|
| 20 |
+
damping=uniform(0.1, 0.8),
|
| 21 |
+
wet_level=1.0,
|
| 22 |
+
dry_level=0.0,
|
| 23 |
+
width=uniform(0.6, 1.0)
|
| 24 |
+
),
|
| 25 |
+
HighpassFilter(cutoff_frequency_hz=uniform(100, 1000)),
|
| 26 |
+
LowpassFilter(cutoff_frequency_hz=uniform(4000, 12000))
|
| 27 |
+
])
|
| 28 |
+
|
| 29 |
+
delay = Pedalboard([
|
| 30 |
+
Delay(
|
| 31 |
+
delay_seconds=uniform(0.05, 0.500),
|
| 32 |
+
feedback=uniform(0.1, 0.5),
|
| 33 |
+
mix=1.0
|
| 34 |
+
),
|
| 35 |
+
Reverb(
|
| 36 |
+
room_size=uniform(0.05, 0.3),
|
| 37 |
+
damping=uniform(0.1, 0.8),
|
| 38 |
+
wet_level=0.2,
|
| 39 |
+
dry_level=0.8,
|
| 40 |
+
width=uniform(0.6, 1.0)
|
| 41 |
+
),
|
| 42 |
+
HighpassFilter(cutoff_frequency_hz=uniform(100, 1000)),
|
| 43 |
+
LowpassFilter(cutoff_frequency_hz=uniform(3000, 10000))
|
| 44 |
+
])
|
| 45 |
+
|
| 46 |
+
effect = uniform(0.1, 0.4) * reverb(audio, sr) + uniform(0.1, 0.4) * delay(audio, sr)
|
| 47 |
+
mix = effect + audio
|
| 48 |
+
|
| 49 |
+
return mix, effect
|
| 50 |
+
|
| 51 |
+
|
| 52 |
+
if __name__ == '__main__':
|
| 53 |
+
argparser = argparse.ArgumentParser(description='Add random reverb and delay effects to an audio file.')
|
| 54 |
+
argparser.add_argument('-i', '--input_folder', type=str, default="train", help='Path to the input audio file.')
|
| 55 |
+
argparser.add_argument('-o', '--output_folder', type=str, default="dataset_train", help='Path to the output audio file.')
|
| 56 |
+
args = argparser.parse_args()
|
| 57 |
+
|
| 58 |
+
index = 1
|
| 59 |
+
sr = 44100
|
| 60 |
+
for file in tqdm(os.listdir(args.input_folder)):
|
| 61 |
+
try:
|
| 62 |
+
audio, _ = librosa.load(os.path.join(args.input_folder, file), sr=sr)
|
| 63 |
+
if len(audio.shape) == 1:
|
| 64 |
+
audio = np.stack([audio, audio], axis=1)
|
| 65 |
+
effect = random_effect(audio.T, sr)
|
| 66 |
+
except:
|
| 67 |
+
print(f"Failed to process file: {file}")
|
| 68 |
+
continue
|
| 69 |
+
|
| 70 |
+
os.makedirs(os.path.join(args.output_folder, str(index)), exist_ok=True)
|
| 71 |
+
|
| 72 |
+
sf.write(os.path.join(args.output_folder, str(index), "mixture.wav"), effect[0].T, sr, subtype='PCM_16')
|
| 73 |
+
sf.write(os.path.join(args.output_folder, str(index), "other.wav"), effect[1].T, sr, subtype='PCM_16')
|
| 74 |
+
sf.write(os.path.join(args.output_folder, str(index), "dry.wav"), audio, sr, subtype='PCM_16')
|
| 75 |
+
|
| 76 |
+
index += 1
|
Dereverb-Echo-MelBand-Sucial/scripts/create_reverb_delay_multi_threads.py
ADDED
|
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import os
|
| 2 |
+
import argparse
|
| 3 |
+
import librosa
|
| 4 |
+
import numpy as np
|
| 5 |
+
import soundfile as sf
|
| 6 |
+
from pedalboard import Pedalboard, Reverb, Delay, HighpassFilter, LowpassFilter
|
| 7 |
+
from random import uniform
|
| 8 |
+
from tqdm import tqdm
|
| 9 |
+
from concurrent.futures import ThreadPoolExecutor
|
| 10 |
+
|
| 11 |
+
def random_effect(audio, sr):
|
| 12 |
+
reverb = Pedalboard([
|
| 13 |
+
Delay(
|
| 14 |
+
delay_seconds=uniform(0.001, 0.100),
|
| 15 |
+
feedback=0.0,
|
| 16 |
+
mix=1.0
|
| 17 |
+
),
|
| 18 |
+
Reverb(
|
| 19 |
+
room_size=uniform(0.7, 1.0),
|
| 20 |
+
damping=uniform(0.7, 1.0),
|
| 21 |
+
wet_level=1.0,
|
| 22 |
+
dry_level=0.0,
|
| 23 |
+
width=uniform(0.7, 1.0)
|
| 24 |
+
),
|
| 25 |
+
HighpassFilter(cutoff_frequency_hz=uniform(100, 800)),
|
| 26 |
+
LowpassFilter(cutoff_frequency_hz=uniform(4000, 15000))
|
| 27 |
+
])
|
| 28 |
+
|
| 29 |
+
effect = uniform(0.3, 0.6) * reverb(audio, sr)
|
| 30 |
+
mix = effect + audio
|
| 31 |
+
|
| 32 |
+
return mix, effect
|
| 33 |
+
|
| 34 |
+
def process_file(file, input_folder, output_folder, index, sr):
|
| 35 |
+
try:
|
| 36 |
+
audio, _ = librosa.load(os.path.join(input_folder, file), sr=sr)
|
| 37 |
+
if len(audio.shape) == 1:
|
| 38 |
+
audio = np.stack([audio, audio], axis=1)
|
| 39 |
+
effect = random_effect(audio.T, sr)
|
| 40 |
+
except Exception as e:
|
| 41 |
+
print(f"Failed to process file: {file}. Error: {e}")
|
| 42 |
+
return False
|
| 43 |
+
|
| 44 |
+
output_path = os.path.join(output_folder, str(index))
|
| 45 |
+
os.makedirs(output_path, exist_ok=True)
|
| 46 |
+
|
| 47 |
+
try:
|
| 48 |
+
sf.write(os.path.join(output_path, "mixture.wav"), effect[0].T, sr, subtype='PCM_16')
|
| 49 |
+
sf.write(os.path.join(output_path, "other.wav"), effect[1].T, sr, subtype='PCM_16')
|
| 50 |
+
sf.write(os.path.join(output_path, "dry.wav"), audio, sr, subtype='PCM_16')
|
| 51 |
+
os.remove(os.path.join(input_folder, file))
|
| 52 |
+
except Exception as e:
|
| 53 |
+
print(f"Failed to save file for {file}. Error: {e}")
|
| 54 |
+
return False
|
| 55 |
+
|
| 56 |
+
return True
|
| 57 |
+
|
| 58 |
+
if __name__ == '__main__':
|
| 59 |
+
argparser = argparse.ArgumentParser(description='Add random reverb and delay effects to audio files using multithreading.')
|
| 60 |
+
argparser.add_argument('-i', '--input_folder', type=str, default="trainset", help='Path to the input folder containing audio files.')
|
| 61 |
+
argparser.add_argument('-o', '--output_folder', type=str, default="train2", help='Path to the output folder for processed audio files.')
|
| 62 |
+
argparser.add_argument('-t', '--threads', type=int, default=32, help='Number of threads to use for processing.')
|
| 63 |
+
args = argparser.parse_args()
|
| 64 |
+
|
| 65 |
+
sr = 44100
|
| 66 |
+
input_files = os.listdir(args.input_folder)
|
| 67 |
+
|
| 68 |
+
with ThreadPoolExecutor(max_workers=args.threads) as executor:
|
| 69 |
+
futures = {executor.submit(process_file, file, args.input_folder, args.output_folder, index, sr): file for index, file in enumerate(input_files, start=1)}
|
| 70 |
+
for future in tqdm(futures, total=len(input_files)):
|
| 71 |
+
future.result()
|
Dereverb-Echo-MelBand-Sucial/scripts/model_fusion.py
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import torch
|
| 2 |
+
|
| 3 |
+
model_1 = torch.load('model_1.ckpt', map_location='cpu')
|
| 4 |
+
model_2 = torch.load('model_2.ckpt', map_location='cpu')
|
| 5 |
+
model_3 = torch.load('model_3.ckpt', map_location='cpu')
|
| 6 |
+
|
| 7 |
+
# Combine the models
|
| 8 |
+
fused_weights = {}
|
| 9 |
+
for key in model_1.keys():
|
| 10 |
+
fused_weights[key] = 0.5 * model_1[key] + 0.25 * model_2[key] + 0.25 * model_3[key]
|
| 11 |
+
|
| 12 |
+
# Save the fused model
|
| 13 |
+
torch.save(fused_weights, 'fused_model.ckpt')
|
Dereverb-MelBand-Anvuew/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3469d6b383b7ae0fff8aa6cb292eedf1eec132d3a960cadeecdec145123566d
|
| 3 |
+
size 438988833
|
Dereverb-MelBand-Anvuew/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.yaml
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 352768
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 256
|
| 5 |
+
hop_length: 512
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.001
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 256
|
| 13 |
+
depth: 6
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
num_bands: 60
|
| 20 |
+
dim_head: 64
|
| 21 |
+
heads: 8
|
| 22 |
+
attn_dropout: 0.1
|
| 23 |
+
ff_dropout: 0.1
|
| 24 |
+
flash_attn: True
|
| 25 |
+
dim_freqs_in: 1025
|
| 26 |
+
sample_rate: 44100 # needed for mel filter bank from librosa
|
| 27 |
+
stft_n_fft: 2048
|
| 28 |
+
stft_hop_length: 512
|
| 29 |
+
stft_win_length: 2048
|
| 30 |
+
stft_normalized: False
|
| 31 |
+
mask_estimator_depth: 2
|
| 32 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 33 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 34 |
+
- 4096
|
| 35 |
+
- 2048
|
| 36 |
+
- 1024
|
| 37 |
+
- 512
|
| 38 |
+
- 256
|
| 39 |
+
multi_stft_hop_size: 147
|
| 40 |
+
multi_stft_normalized: False
|
| 41 |
+
|
| 42 |
+
training:
|
| 43 |
+
batch_size: 1
|
| 44 |
+
gradient_accumulation_steps: 1
|
| 45 |
+
grad_clip: 0
|
| 46 |
+
instruments:
|
| 47 |
+
- noreverb
|
| 48 |
+
- reverb
|
| 49 |
+
lr: 5.0e-05
|
| 50 |
+
patience: 2
|
| 51 |
+
reduce_factor: 0.95
|
| 52 |
+
target_instrument: noreverb
|
| 53 |
+
num_epochs: 1000
|
| 54 |
+
num_steps: 40000
|
| 55 |
+
q: 0.95
|
| 56 |
+
coarse_loss_clip: true
|
| 57 |
+
ema_momentum: 0.999
|
| 58 |
+
optimizer: adam
|
| 59 |
+
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
|
| 60 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 61 |
+
|
| 62 |
+
augmentations:
|
| 63 |
+
enable: true # enable or disable all augmentations (to fast disable if needed)
|
| 64 |
+
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
|
| 65 |
+
loudness_min: 0.5
|
| 66 |
+
loudness_max: 1.5
|
| 67 |
+
mixup: false # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
|
| 68 |
+
mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
|
| 69 |
+
- 0.2
|
| 70 |
+
- 0.02
|
| 71 |
+
mixup_loudness_min: 0.5
|
| 72 |
+
mixup_loudness_max: 1.5
|
| 73 |
+
|
| 74 |
+
inference:
|
| 75 |
+
batch_size: 1
|
| 76 |
+
dim_t: 256
|
| 77 |
+
num_overlap: 4
|
Dereverb-MelBand-Anvuew/archive only/8_512_12/deverb_mel_band_roformer_8_512dim_12depth.yaml
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 352768
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 801
|
| 5 |
+
hop_length: 441
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.001
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 512
|
| 13 |
+
depth: 12
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
num_bands: 60
|
| 20 |
+
dim_head: 64
|
| 21 |
+
heads: 8
|
| 22 |
+
attn_dropout: 0.1
|
| 23 |
+
ff_dropout: 0.1
|
| 24 |
+
flash_attn: True
|
| 25 |
+
dim_freqs_in: 1025
|
| 26 |
+
sample_rate: 44100 # needed for mel filter bank from librosa
|
| 27 |
+
stft_n_fft: 2048
|
| 28 |
+
stft_hop_length: 512
|
| 29 |
+
stft_win_length: 2048
|
| 30 |
+
stft_normalized: False
|
| 31 |
+
mask_estimator_depth: 2
|
| 32 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 33 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 34 |
+
- 4096
|
| 35 |
+
- 2048
|
| 36 |
+
- 1024
|
| 37 |
+
- 512
|
| 38 |
+
- 256
|
| 39 |
+
multi_stft_hop_size: 147
|
| 40 |
+
multi_stft_normalized: False
|
| 41 |
+
|
| 42 |
+
training:
|
| 43 |
+
batch_size: 2
|
| 44 |
+
gradient_accumulation_steps: 1
|
| 45 |
+
grad_clip: 0
|
| 46 |
+
instruments:
|
| 47 |
+
- noreverb
|
| 48 |
+
- reverb
|
| 49 |
+
lr: 5.0e-05
|
| 50 |
+
patience: 2
|
| 51 |
+
reduce_factor: 0.95
|
| 52 |
+
target_instrument: noreverb
|
| 53 |
+
num_epochs: 1000
|
| 54 |
+
num_steps: 1900
|
| 55 |
+
q: 0.95
|
| 56 |
+
coarse_loss_clip: true
|
| 57 |
+
ema_momentum: 0.999
|
| 58 |
+
optimizer: adam
|
| 59 |
+
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
|
| 60 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 61 |
+
|
| 62 |
+
augmentations:
|
| 63 |
+
enable: true # enable or disable all augmentations (to fast disable if needed)
|
| 64 |
+
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
|
| 65 |
+
loudness_min: 0.5
|
| 66 |
+
loudness_max: 1.5
|
| 67 |
+
mixup: false # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
|
| 68 |
+
mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
|
| 69 |
+
- 0.2
|
| 70 |
+
- 0.02
|
| 71 |
+
mixup_loudness_min: 0.5
|
| 72 |
+
mixup_loudness_max: 1.5
|
| 73 |
+
|
| 74 |
+
inference:
|
| 75 |
+
batch_size: 1
|
| 76 |
+
dim_t: 801
|
| 77 |
+
num_overlap: 4
|
HuggingFace_Repos/anvuew/BS-RoFormer/.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
HuggingFace_Repos/anvuew/BS-RoFormer/README.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gpl-3.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
dataset by [bascurtiz](https://github.com/bascurtiz)
|
HuggingFace_Repos/anvuew/BS-RoFormer/bs_roformer_anvuew_sdr_12.45.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d0f72ad0ac4154f5f4c3cdf230880bbb9a5bd01dc610241a2abfe787f0d7784
|
| 3 |
+
size 204485563
|
HuggingFace_Repos/anvuew/BS-RoFormer/config.yaml
ADDED
|
@@ -0,0 +1,129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 960000
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 801 # don't work (use in model)
|
| 5 |
+
hop_length: 441 # don't work (use in model)
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.0001
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 256
|
| 13 |
+
depth: 12
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
freqs_per_bands: !!python/tuple
|
| 20 |
+
- 2
|
| 21 |
+
- 2
|
| 22 |
+
- 2
|
| 23 |
+
- 2
|
| 24 |
+
- 2
|
| 25 |
+
- 2
|
| 26 |
+
- 2
|
| 27 |
+
- 2
|
| 28 |
+
- 2
|
| 29 |
+
- 2
|
| 30 |
+
- 2
|
| 31 |
+
- 2
|
| 32 |
+
- 2
|
| 33 |
+
- 2
|
| 34 |
+
- 2
|
| 35 |
+
- 2
|
| 36 |
+
- 2
|
| 37 |
+
- 2
|
| 38 |
+
- 2
|
| 39 |
+
- 2
|
| 40 |
+
- 2
|
| 41 |
+
- 2
|
| 42 |
+
- 2
|
| 43 |
+
- 2
|
| 44 |
+
- 4
|
| 45 |
+
- 4
|
| 46 |
+
- 4
|
| 47 |
+
- 4
|
| 48 |
+
- 4
|
| 49 |
+
- 4
|
| 50 |
+
- 4
|
| 51 |
+
- 4
|
| 52 |
+
- 4
|
| 53 |
+
- 4
|
| 54 |
+
- 4
|
| 55 |
+
- 4
|
| 56 |
+
- 12
|
| 57 |
+
- 12
|
| 58 |
+
- 12
|
| 59 |
+
- 12
|
| 60 |
+
- 12
|
| 61 |
+
- 12
|
| 62 |
+
- 12
|
| 63 |
+
- 12
|
| 64 |
+
- 24
|
| 65 |
+
- 24
|
| 66 |
+
- 24
|
| 67 |
+
- 24
|
| 68 |
+
- 24
|
| 69 |
+
- 24
|
| 70 |
+
- 24
|
| 71 |
+
- 24
|
| 72 |
+
- 48
|
| 73 |
+
- 48
|
| 74 |
+
- 48
|
| 75 |
+
- 48
|
| 76 |
+
- 48
|
| 77 |
+
- 48
|
| 78 |
+
- 48
|
| 79 |
+
- 48
|
| 80 |
+
- 128
|
| 81 |
+
- 129
|
| 82 |
+
dim_head: 64
|
| 83 |
+
heads: 8
|
| 84 |
+
attn_dropout: 0.0
|
| 85 |
+
ff_dropout: 0.0
|
| 86 |
+
flash_attn: true
|
| 87 |
+
dim_freqs_in: 1025
|
| 88 |
+
stft_n_fft: 2048
|
| 89 |
+
stft_hop_length: 512
|
| 90 |
+
stft_win_length: 2048
|
| 91 |
+
stft_normalized: false
|
| 92 |
+
mask_estimator_depth: 2
|
| 93 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 94 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 95 |
+
- 4096
|
| 96 |
+
- 2048
|
| 97 |
+
- 1024
|
| 98 |
+
- 512
|
| 99 |
+
- 256
|
| 100 |
+
multi_stft_hop_size: 147
|
| 101 |
+
multi_stft_normalized: False
|
| 102 |
+
mlp_expansion_factor: 4
|
| 103 |
+
use_torch_checkpoint: True
|
| 104 |
+
skip_connection: False
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
training:
|
| 108 |
+
batch_size: 1
|
| 109 |
+
gradient_accumulation_steps: 1
|
| 110 |
+
grad_clip: 0
|
| 111 |
+
instruments: ['vocals', 'instrument']
|
| 112 |
+
lr: 1.0e-5
|
| 113 |
+
patience: 5
|
| 114 |
+
reduce_factor: 0.9
|
| 115 |
+
target_instrument: vocals
|
| 116 |
+
num_epochs: 1000
|
| 117 |
+
num_steps: 1000
|
| 118 |
+
q: 0.95
|
| 119 |
+
coarse_loss_clip: true
|
| 120 |
+
ema_momentum: 0.999
|
| 121 |
+
optimizer: adam
|
| 122 |
+
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
|
| 123 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 124 |
+
|
| 125 |
+
|
| 126 |
+
inference:
|
| 127 |
+
batch_size: 2
|
| 128 |
+
dim_t: 1876
|
| 129 |
+
num_overlap: 4
|
HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/README.md
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gpl-3.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
专用于人声的去混响模型,适用于 https://github.com/ZFTurbo/Music-Source-Separation-Training
|
| 6 |
+
|
| 7 |
+
微调自[kim的人声乐器分离模型](https://huggingface.co/KimberleyJSN/melbandroformer),所以保留了一些分离人声与乐器的能力
|
| 8 |
+
|
| 9 |
+
训练数据的干声为单声道单人演唱/说话,所以模型会尽可能使分离后的干声为单声道,导致在分离一些人声为立体声或者经过叠轨等处理的音频时可能会出问题,`dereverb_mel_band_roformer_less_aggressive_anvuew_sdr_18.8050.ckpt`是使用为这些情况的样本挑出来的`dereverb_mel_band_roformer_anvuew_sdr_19.1729.ckpt`训练中期的权重
|
| 10 |
+
|
| 11 |
+
`dereverb_mel_band_roformer_anvuew_sdr_19.1729.ckpt`与`dereverb_mel_band_roformer_less_aggressive_anvuew_sdr_18.8050.ckpt`两个权重由于训练代码[bug](https://github.com/ZFTurbo/Music-Source-Separation-Training/commit/0ca5691f22ea71d9afe297926d6e1517cdb38e55),在训练时混响与人声实际并未对齐,所以这两个模型表现出一定程度的去除人声与乐器分离后的残留(通常是弦乐)以及部分和声(非中央声道)的能力
|
| 12 |
+
|
| 13 |
+
`dereverb_mel_band_roformer_mono_anvuew_sdr_20.4029.ckpt`则因修复了对齐bug,去混响的能力更强了,但是去人声乐器分离残留以及和声的能力下降了
|
| 14 |
+
|
| 15 |
+
|
HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/archive only/8_256_6/deverb_mel_band_roformer_8_256dim_6depth.yaml
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 352768
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 256
|
| 5 |
+
hop_length: 512
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.001
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 256
|
| 13 |
+
depth: 6
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
num_bands: 60
|
| 20 |
+
dim_head: 64
|
| 21 |
+
heads: 8
|
| 22 |
+
attn_dropout: 0.1
|
| 23 |
+
ff_dropout: 0.1
|
| 24 |
+
flash_attn: True
|
| 25 |
+
dim_freqs_in: 1025
|
| 26 |
+
sample_rate: 44100 # needed for mel filter bank from librosa
|
| 27 |
+
stft_n_fft: 2048
|
| 28 |
+
stft_hop_length: 512
|
| 29 |
+
stft_win_length: 2048
|
| 30 |
+
stft_normalized: False
|
| 31 |
+
mask_estimator_depth: 2
|
| 32 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 33 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 34 |
+
- 4096
|
| 35 |
+
- 2048
|
| 36 |
+
- 1024
|
| 37 |
+
- 512
|
| 38 |
+
- 256
|
| 39 |
+
multi_stft_hop_size: 147
|
| 40 |
+
multi_stft_normalized: False
|
| 41 |
+
|
| 42 |
+
training:
|
| 43 |
+
batch_size: 1
|
| 44 |
+
gradient_accumulation_steps: 1
|
| 45 |
+
grad_clip: 0
|
| 46 |
+
instruments:
|
| 47 |
+
- noreverb
|
| 48 |
+
- reverb
|
| 49 |
+
lr: 5.0e-05
|
| 50 |
+
patience: 2
|
| 51 |
+
reduce_factor: 0.95
|
| 52 |
+
target_instrument: noreverb
|
| 53 |
+
num_epochs: 1000
|
| 54 |
+
num_steps: 40000
|
| 55 |
+
q: 0.95
|
| 56 |
+
coarse_loss_clip: true
|
| 57 |
+
ema_momentum: 0.999
|
| 58 |
+
optimizer: adam
|
| 59 |
+
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
|
| 60 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 61 |
+
|
| 62 |
+
augmentations:
|
| 63 |
+
enable: true # enable or disable all augmentations (to fast disable if needed)
|
| 64 |
+
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
|
| 65 |
+
loudness_min: 0.5
|
| 66 |
+
loudness_max: 1.5
|
| 67 |
+
mixup: false # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
|
| 68 |
+
mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
|
| 69 |
+
- 0.2
|
| 70 |
+
- 0.02
|
| 71 |
+
mixup_loudness_min: 0.5
|
| 72 |
+
mixup_loudness_max: 1.5
|
| 73 |
+
|
| 74 |
+
inference:
|
| 75 |
+
batch_size: 1
|
| 76 |
+
dim_t: 256
|
| 77 |
+
num_overlap: 4
|
HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/archive only/8_512_12/deverb_mel_band_roformer_8_512dim_12depth.yaml
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 352768
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 801
|
| 5 |
+
hop_length: 441
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.001
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 512
|
| 13 |
+
depth: 12
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
num_bands: 60
|
| 20 |
+
dim_head: 64
|
| 21 |
+
heads: 8
|
| 22 |
+
attn_dropout: 0.1
|
| 23 |
+
ff_dropout: 0.1
|
| 24 |
+
flash_attn: True
|
| 25 |
+
dim_freqs_in: 1025
|
| 26 |
+
sample_rate: 44100 # needed for mel filter bank from librosa
|
| 27 |
+
stft_n_fft: 2048
|
| 28 |
+
stft_hop_length: 512
|
| 29 |
+
stft_win_length: 2048
|
| 30 |
+
stft_normalized: False
|
| 31 |
+
mask_estimator_depth: 2
|
| 32 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 33 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 34 |
+
- 4096
|
| 35 |
+
- 2048
|
| 36 |
+
- 1024
|
| 37 |
+
- 512
|
| 38 |
+
- 256
|
| 39 |
+
multi_stft_hop_size: 147
|
| 40 |
+
multi_stft_normalized: False
|
| 41 |
+
|
| 42 |
+
training:
|
| 43 |
+
batch_size: 2
|
| 44 |
+
gradient_accumulation_steps: 1
|
| 45 |
+
grad_clip: 0
|
| 46 |
+
instruments:
|
| 47 |
+
- noreverb
|
| 48 |
+
- reverb
|
| 49 |
+
lr: 5.0e-05
|
| 50 |
+
patience: 2
|
| 51 |
+
reduce_factor: 0.95
|
| 52 |
+
target_instrument: noreverb
|
| 53 |
+
num_epochs: 1000
|
| 54 |
+
num_steps: 1900
|
| 55 |
+
q: 0.95
|
| 56 |
+
coarse_loss_clip: true
|
| 57 |
+
ema_momentum: 0.999
|
| 58 |
+
optimizer: adam
|
| 59 |
+
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
|
| 60 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 61 |
+
|
| 62 |
+
augmentations:
|
| 63 |
+
enable: true # enable or disable all augmentations (to fast disable if needed)
|
| 64 |
+
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
|
| 65 |
+
loudness_min: 0.5
|
| 66 |
+
loudness_max: 1.5
|
| 67 |
+
mixup: false # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
|
| 68 |
+
mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
|
| 69 |
+
- 0.2
|
| 70 |
+
- 0.02
|
| 71 |
+
mixup_loudness_min: 0.5
|
| 72 |
+
mixup_loudness_max: 1.5
|
| 73 |
+
|
| 74 |
+
inference:
|
| 75 |
+
batch_size: 1
|
| 76 |
+
dim_t: 801
|
| 77 |
+
num_overlap: 4
|
HuggingFace_Repos/anvuew/dereverb_mel_band_roformer/dereverb_mel_band_roformer_anvuew.yaml
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 352800
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 256
|
| 5 |
+
hop_length: 441
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.000
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 384
|
| 13 |
+
depth: 6
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
num_bands: 60
|
| 19 |
+
dim_head: 64
|
| 20 |
+
heads: 8
|
| 21 |
+
attn_dropout: 0
|
| 22 |
+
ff_dropout: 0
|
| 23 |
+
flash_attn: True
|
| 24 |
+
dim_freqs_in: 1025
|
| 25 |
+
sample_rate: 44100 # needed for mel filter bank from librosa
|
| 26 |
+
stft_n_fft: 2048
|
| 27 |
+
stft_hop_length: 441
|
| 28 |
+
stft_win_length: 2048
|
| 29 |
+
stft_normalized: False
|
| 30 |
+
mask_estimator_depth: 2
|
| 31 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 32 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 33 |
+
- 4096
|
| 34 |
+
- 2048
|
| 35 |
+
- 1024
|
| 36 |
+
- 512
|
| 37 |
+
- 256
|
| 38 |
+
multi_stft_hop_size: 147
|
| 39 |
+
multi_stft_normalized: False
|
| 40 |
+
|
| 41 |
+
training:
|
| 42 |
+
batch_size: 3
|
| 43 |
+
gradient_accumulation_steps: 1
|
| 44 |
+
grad_clip: 0
|
| 45 |
+
instruments:
|
| 46 |
+
- noreverb
|
| 47 |
+
- reverb
|
| 48 |
+
lr: 5.0e-05
|
| 49 |
+
patience: 2
|
| 50 |
+
reduce_factor: 0.95
|
| 51 |
+
target_instrument: noreverb
|
| 52 |
+
num_epochs: 1000
|
| 53 |
+
num_steps: 4000
|
| 54 |
+
q: 0.95
|
| 55 |
+
coarse_loss_clip: false
|
| 56 |
+
ema_momentum: 0.999
|
| 57 |
+
optimizer: adamw
|
| 58 |
+
other_fix: true # it's needed for checking on multisong dataset if other is actually instrumental
|
| 59 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 60 |
+
|
| 61 |
+
augmentations:
|
| 62 |
+
enable: true # enable or disable all augmentations (to fast disable if needed)
|
| 63 |
+
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
|
| 64 |
+
loudness_min: 0.1
|
| 65 |
+
loudness_max: 1.0
|
| 66 |
+
mixup: false # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
|
| 67 |
+
mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
|
| 68 |
+
- 0.2
|
| 69 |
+
- 0.02
|
| 70 |
+
mixup_loudness_min: 0.5
|
| 71 |
+
mixup_loudness_max: 1.5
|
| 72 |
+
|
| 73 |
+
inference:
|
| 74 |
+
batch_size: 1
|
| 75 |
+
dim_t: 801
|
| 76 |
+
num_overlap: 2
|
HuggingFace_Repos/anvuew/karaoke_bs_roformer/.gitattributes
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz filter=lfs diff=lfs merge=lfs -text
|
| 33 |
+
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
+
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
+
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
HuggingFace_Repos/anvuew/karaoke_bs_roformer/README.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gpl-3.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
dataset by [becruily](https://huggingface.co/becruily)
|
HuggingFace_Repos/anvuew/karaoke_bs_roformer/karaoke_bs_roformer_anvuew.ckpt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:206d04757cb5f75ca3b55f8a0a48f5c26aa2351d4ff3c7adbfc9affa30ea3ae4
|
| 3 |
+
size 204486925
|
HuggingFace_Repos/anvuew/karaoke_bs_roformer/karaoke_bs_roformer_anvuew.yaml
ADDED
|
@@ -0,0 +1,131 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
audio:
|
| 2 |
+
chunk_size: 640000
|
| 3 |
+
dim_f: 1024
|
| 4 |
+
dim_t: 801 # don't work (use in model)
|
| 5 |
+
hop_length: 441 # don't work (use in model)
|
| 6 |
+
n_fft: 2048
|
| 7 |
+
num_channels: 2
|
| 8 |
+
sample_rate: 44100
|
| 9 |
+
min_mean_abs: 0.000
|
| 10 |
+
|
| 11 |
+
model:
|
| 12 |
+
dim: 256
|
| 13 |
+
depth: 12
|
| 14 |
+
stereo: true
|
| 15 |
+
num_stems: 1
|
| 16 |
+
time_transformer_depth: 1
|
| 17 |
+
freq_transformer_depth: 1
|
| 18 |
+
linear_transformer_depth: 0
|
| 19 |
+
freqs_per_bands: !!python/tuple
|
| 20 |
+
- 2
|
| 21 |
+
- 2
|
| 22 |
+
- 2
|
| 23 |
+
- 2
|
| 24 |
+
- 2
|
| 25 |
+
- 2
|
| 26 |
+
- 2
|
| 27 |
+
- 2
|
| 28 |
+
- 2
|
| 29 |
+
- 2
|
| 30 |
+
- 2
|
| 31 |
+
- 2
|
| 32 |
+
- 2
|
| 33 |
+
- 2
|
| 34 |
+
- 2
|
| 35 |
+
- 2
|
| 36 |
+
- 2
|
| 37 |
+
- 2
|
| 38 |
+
- 2
|
| 39 |
+
- 2
|
| 40 |
+
- 2
|
| 41 |
+
- 2
|
| 42 |
+
- 2
|
| 43 |
+
- 2
|
| 44 |
+
- 4
|
| 45 |
+
- 4
|
| 46 |
+
- 4
|
| 47 |
+
- 4
|
| 48 |
+
- 4
|
| 49 |
+
- 4
|
| 50 |
+
- 4
|
| 51 |
+
- 4
|
| 52 |
+
- 4
|
| 53 |
+
- 4
|
| 54 |
+
- 4
|
| 55 |
+
- 4
|
| 56 |
+
- 12
|
| 57 |
+
- 12
|
| 58 |
+
- 12
|
| 59 |
+
- 12
|
| 60 |
+
- 12
|
| 61 |
+
- 12
|
| 62 |
+
- 12
|
| 63 |
+
- 12
|
| 64 |
+
- 24
|
| 65 |
+
- 24
|
| 66 |
+
- 24
|
| 67 |
+
- 24
|
| 68 |
+
- 24
|
| 69 |
+
- 24
|
| 70 |
+
- 24
|
| 71 |
+
- 24
|
| 72 |
+
- 48
|
| 73 |
+
- 48
|
| 74 |
+
- 48
|
| 75 |
+
- 48
|
| 76 |
+
- 48
|
| 77 |
+
- 48
|
| 78 |
+
- 48
|
| 79 |
+
- 48
|
| 80 |
+
- 128
|
| 81 |
+
- 129
|
| 82 |
+
dim_head: 64
|
| 83 |
+
heads: 8
|
| 84 |
+
attn_dropout: 0.0
|
| 85 |
+
ff_dropout: 0.0
|
| 86 |
+
flash_attn: true
|
| 87 |
+
dim_freqs_in: 1025
|
| 88 |
+
stft_n_fft: 2048
|
| 89 |
+
stft_hop_length: 512
|
| 90 |
+
stft_win_length: 2048
|
| 91 |
+
stft_normalized: false
|
| 92 |
+
mask_estimator_depth: 2
|
| 93 |
+
multi_stft_resolution_loss_weight: 1.0
|
| 94 |
+
multi_stft_resolutions_window_sizes: !!python/tuple
|
| 95 |
+
- 4096
|
| 96 |
+
- 2048
|
| 97 |
+
- 1024
|
| 98 |
+
- 512
|
| 99 |
+
- 256
|
| 100 |
+
multi_stft_hop_size: 147
|
| 101 |
+
multi_stft_normalized: False
|
| 102 |
+
mlp_expansion_factor: 4
|
| 103 |
+
use_torch_checkpoint: True
|
| 104 |
+
skip_connection: False
|
| 105 |
+
|
| 106 |
+
|
| 107 |
+
training:
|
| 108 |
+
batch_size: 1
|
| 109 |
+
gradient_accumulation_steps: 1
|
| 110 |
+
grad_clip: 0
|
| 111 |
+
instruments: ['Vocals', 'Instrumental']
|
| 112 |
+
lr: 5.0e-5
|
| 113 |
+
patience: 7
|
| 114 |
+
reduce_factor: 0.75
|
| 115 |
+
target_instrument: Vocals
|
| 116 |
+
num_epochs: 1000
|
| 117 |
+
num_steps: 1000
|
| 118 |
+
q: 0.95
|
| 119 |
+
coarse_loss_clip: true
|
| 120 |
+
ema_momentum: 0.999
|
| 121 |
+
optimizer: adam
|
| 122 |
+
other_fix: False # it's needed for checking on multisong dataset if other is actually instrumental
|
| 123 |
+
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
|
| 124 |
+
|
| 125 |
+
|
| 126 |
+
|
| 127 |
+
|
| 128 |
+
inference:
|
| 129 |
+
batch_size: 2
|
| 130 |
+
dim_t: 1251
|
| 131 |
+
num_overlap: 4
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/.~lock.all_metrics.csv#
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
,carson,carson-MS-7B86,01.10.2023 22:35,file:///home/carson/.config/libreoffice/4;
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/all_metrics.csv
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
,input_si_sdr,input_sdr,input_sir,input_sar,input_stoi,si_sdr,sdr,sir,sar,stoi,mix_path
|
| 2 |
+
0,0.0911402702331543,0.11333474572369884,0.11333570336043364,72.0603379655421,0.6399578763503541,15.064705848693848,15.721711668127007,23.45321172828264,16.616873320628407,0.786226926844334,../data/MedleyVox/duet/CatMartino_IPromise/seg_1/mix/CatMartino_IPromise - seg_1.wav
|
| 3 |
+
1,0.02406684309244156,0.2720073709833361,0.2720112399176431,63.6579561715186,0.8001933107257376,-0.020107148215174675,2.3113227359507356,3.515217767392589,10.170321009767745,0.7535998004890577,../data/MedleyVox/duet/CatMartino_IPromise/seg_17/mix/CatMartino_IPromise - seg_17.wav
|
| 4 |
+
2,0.023504503071308136,0.26802173592922623,0.2680257247729072,63.525063512012906,0.8064488134774055,-0.06741892546415329,2.507732191724367,3.8045505193249385,10.04598635996416,0.7593386231763503,../data/MedleyVox/duet/CatMartino_IPromise/seg_21/mix/CatMartino_IPromise - seg_21.wav
|
| 5 |
+
3,0.0038509368896484375,0.02030408099396208,0.02030560562826178,72.79871692374104,0.5561395325418792,15.383918285369873,15.737125832356575,24.282791715101162,16.440832010953336,0.6852088280754502,../data/MedleyVox/duet/CatMartino_IPromise/seg_3/mix/CatMartino_IPromise - seg_3.wav
|
| 6 |
+
4,0.02143096923828125,0.07497634692502197,0.07497716754543271,73.59395962015613,0.5809720090023308,18.3359694480896,18.943418834251837,26.891610730720096,19.71975365185834,0.7372336291827839,../data/MedleyVox/duet/CatMartino_IPromise/seg_4/mix/CatMartino_IPromise - seg_4.wav
|
| 7 |
+
5,0.08760905265808105,0.13416892932315339,0.13416982711127678,70.56421253110653,0.8152177005156398,15.072598934173584,16.203999512297298,21.880150457227387,17.806424371854582,0.8673778392006342,../data/MedleyVox/duet/CatMartino_IPromise/seg_7/mix/CatMartino_IPromise - seg_7.wav
|
| 8 |
+
6,0.007455706596374512,0.018030733424175535,0.018031698569134134,69.55197961211707,0.7500427475771034,14.072438716888428,14.738300718601948,21.78203148325104,15.735985035398913,0.8794454459382519,../data/MedleyVox/duet/Cayetana_MissThing/seg_1/mix/Cayetana_MissThing - seg_1.wav
|
| 9 |
+
7,0.02685832977294922,0.03738885653335555,0.03738955054481585,71.38825034540149,0.6337365420795307,22.054887771606445,22.60662895609594,31.919236416591158,23.15909815031467,0.7678895779600658,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_45/mix/CelestialShore_DieForUs - seg_45.wav
|
| 10 |
+
8,0.05670976638793945,0.07074082917372415,0.07074144009968053,71.73390521848195,0.6185596781047933,20.796342849731445,21.4287955239168,30.2553888493433,22.052916149552015,0.7724372904050387,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_46/mix/CelestialShore_DieForUs - seg_46.wav
|
| 11 |
+
9,0.14332962036132812,0.16262581377419005,0.16262642313589504,71.86363741388934,0.6311706650878312,21.050506591796875,21.42940083451248,31.85931459208959,21.854635591169455,0.7549824236469984,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_47/mix/CelestialShore_DieForUs - seg_47.wav
|
| 12 |
+
10,0.128265380859375,0.15324910005827275,0.15324981439480334,71.23255351709889,0.6154861933701743,18.682246208190918,19.852671139114562,26.24738617896979,21.231095514496445,0.7459573741576933,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_48/mix/CelestialShore_DieForUs - seg_48.wav
|
| 13 |
+
11,0.05758732557296753,0.077363043101559,0.07736349149066901,72.95766862598383,0.6259839452526627,18.299739837646484,19.68875005442171,24.167130644669122,21.621813677404358,0.8269215766554261,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_49/mix/CelestialShore_DieForUs - seg_49.wav
|
| 14 |
+
12,0.019161254167556763,0.04505861820086385,0.04505906214744386,72.93893489031318,0.6886667461155986,16.1737003326416,17.901584446738386,21.78600954508526,20.443524361489057,0.8197986735845563,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_50/mix/CelestialShore_DieForUs - seg_50.wav
|
| 15 |
+
13,-0.017694473266601562,0.0046321481555779975,0.00463247629462793,74.25089108994348,0.635374454270375,5.39709210395813,6.9427699587649405,10.314802780070849,10.151922793147993,0.7573186036038438,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_51/mix/CelestialShore_DieForUs - seg_51.wav
|
| 16 |
+
14,-0.00019288063049316406,0.005274560710137877,0.0052754997584760766,70.16690266362302,0.6232696854447997,19.648597717285156,20.21806099973253,28.46668275168083,20.92981730152865,0.7516888201749573,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_52/mix/CelestialShore_DieForUs - seg_52.wav
|
| 17 |
+
15,0.01592230796813965,0.025396623848236155,0.025397413560217963,70.82578262116812,0.6330517208512643,22.348715782165527,22.816088700138536,32.36840619464648,23.353991968400457,0.7241502398664481,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_53/mix/CelestialShore_DieForUs - seg_53.wav
|
| 18 |
+
16,0.003073453903198242,0.033736624601256926,0.033737051234212556,73.53131209257162,0.6276588895157553,18.52164936065674,19.85018254937779,24.813957542509133,21.556240266013447,0.8036713182947912,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_54/mix/CelestialShore_DieForUs - seg_54.wav
|
| 19 |
+
17,0.011256307363510132,0.02321607777696777,0.0232165199284165,72.97352955149086,0.6509873487434956,16.299922943115234,17.652385603934047,22.273909002135262,19.556394559500987,0.809833185660897,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_55/mix/CelestialShore_DieForUs - seg_55.wav
|
| 20 |
+
18,-0.20395523309707642,-0.17234465115083925,-0.17234428701259386,73.74059066668164,0.6297016626189389,16.537992477416992,17.20660816665272,24.618686087466358,18.091599431352144,0.7437726891726555,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_56/mix/CelestialShore_DieForUs - seg_56.wav
|
| 21 |
+
19,-0.013024091720581055,0.04038188597472403,0.04038262248236868,71.23421055008231,0.6422089449986654,16.5218186378479,18.0772363595566,23.224643004369007,19.890008371372808,0.8601791425499175,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_1/mix/ClaraBerryAndWooldog_Boys - seg_1.wav
|
| 22 |
+
20,-0.0430682897567749,0.02098904110130828,0.020989503678424892,73.01358518576015,0.6004151029924082,13.249131679534912,15.602833915547102,19.7363751054923,18.740614789237227,0.8243840842086296,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_4/mix/ClaraBerryAndWooldog_Boys - seg_4.wav
|
| 23 |
+
21,0.005000650882720947,0.06638932945126408,0.06639005426505118,70.88366644865249,0.5542043543510237,12.635054111480713,13.516495577424081,19.468788731723624,14.880350175371287,0.7799418765644617,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_7/mix/ClaraBerryAndWooldog_Boys - seg_7.wav
|
| 24 |
+
22,0.09175297617912292,0.15464052461563546,0.15464089564175149,73.77624837319533,0.598772825417861,16.2485408782959,16.90885271979107,25.65541937289863,17.76630946574916,0.7755545546360976,../data/MedleyVox/duet/FacesOnFilm_WaitingForGa/seg_2/mix/FacesOnFilm_WaitingForGa - seg_2.wav
|
| 25 |
+
23,0.0015957951545715332,0.06626148109154995,0.06626181714834178,74.32902456467983,0.7016102840806384,15.02765417098999,16.159035924338454,20.933781817674898,17.967644453371612,0.8423310915450568,../data/MedleyVox/duet/FamilyBand_Again/seg_1/mix/FamilyBand_Again - seg_1.wav
|
| 26 |
+
24,0.02108180522918701,0.07702722333964185,0.07702755382179804,74.34959429172716,0.6719157957177997,16.179515838623047,17.468981712297413,22.645810248540972,19.06835543986324,0.7918739004245372,../data/MedleyVox/duet/FamilyBand_Again/seg_2/mix/FamilyBand_Again - seg_2.wav
|
| 27 |
+
25,0.0210113525390625,0.44999612847186565,0.44999652144997726,74.61917157860216,0.5805111802982735,12.927275657653809,16.289788577343007,18.52990601344194,21.36340245510996,0.7509054192736817,../data/MedleyVox/duet/FamilyBand_Again/seg_6/mix/FamilyBand_Again - seg_6.wav
|
| 28 |
+
26,0.07987546920776367,0.13272754482617088,0.13272928396738415,69.83361102952001,0.6816841174395663,13.935612678527832,15.01693113581746,21.280237634025642,16.376275935368984,0.8236018744669136,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_1/mix/FilthyBird_IdLikeToKnow - seg_1.wav
|
| 29 |
+
27,0.004979491233825684,0.011056028115652039,0.011057087103157948,69.20021707551822,0.6649619360356808,19.533611297607422,20.267953741002437,27.701597084384716,21.143146304117614,0.8537532829383045,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_3/mix/FilthyBird_IdLikeToKnow - seg_3.wav
|
| 30 |
+
28,-0.02172267436981201,-0.01886554784865746,-0.018864495288558936,69.34038650573328,0.6960364677432993,17.261977195739746,17.636927828089505,27.249503047893555,18.154920651068643,0.8144059119128467,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_4/mix/FilthyBird_IdLikeToKnow - seg_4.wav
|
| 31 |
+
29,-0.02795553207397461,0.19674289669623057,0.19674333759331475,74.84132036495171,0.6265701085920463,12.997706413269043,13.611380435620491,21.465391951301797,14.431234891438464,0.7482111423515723,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_11/mix/HeladoNegro_MitadDelMundo - seg_11.wav
|
| 32 |
+
30,-2.0922908782958984,-0.7448119301560405,-0.7448115538031477,75.3771459414981,0.6339736535992503,11.389642238616943,12.371295986388002,19.80448254949908,13.621993479900416,0.7732099965197361,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_13/mix/HeladoNegro_MitadDelMundo - seg_13.wav
|
| 33 |
+
31,-0.0640568733215332,0.18826454995281483,0.18826492101386982,74.51669068820232,0.6173298481221636,11.916536331176758,12.578837380096722,19.726239071302526,13.560473379535393,0.7578102498259681,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_14/mix/HeladoNegro_MitadDelMundo - seg_14.wav
|
| 34 |
+
32,0.031224727630615234,0.2817567609715672,0.2817577106486384,71.29556470748452,0.6797899387153046,11.72071361541748,12.862037616691953,18.356897791192704,14.40251155206149,0.8648858307917939,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_16/mix/HeladoNegro_MitadDelMundo - seg_16.wav
|
| 35 |
+
33,-0.29101765155792236,-0.13833882530884156,-0.13833836703988234,73.44498086768832,0.5691573747245798,11.118027210235596,11.890470450454284,18.12829352827788,13.314634424739223,0.6970895882742073,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_17/mix/HeladoNegro_MitadDelMundo - seg_17.wav
|
| 36 |
+
34,0.02745962142944336,0.09716946750039268,0.09717043462993269,71.83073233577244,0.6162581878474078,11.833811283111572,12.116783113903214,21.407837997767192,12.761269858080432,0.7401380382506599,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_18/mix/HeladoNegro_MitadDelMundo - seg_18.wav
|
| 37 |
+
35,-0.26619768142700195,-0.213581786451837,-0.21358096418494954,71.96222733630219,0.5950415002528179,11.728025436401367,12.033227622062457,20.653572407676734,12.885787486923437,0.7474663681666938,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_19/mix/HeladoNegro_MitadDelMundo - seg_19.wav
|
| 38 |
+
36,0.08208179473876953,0.1343153237494228,0.13431603118546098,72.45404927883251,0.5775131774705757,10.955091953277588,11.723838964179134,19.895606859373842,12.908362924388694,0.7181244679719557,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_20/mix/HeladoNegro_MitadDelMundo - seg_20.wav
|
| 39 |
+
37,-0.22986102104187012,0.029813558818715613,0.029814220321483287,72.73544176790732,0.7634997694258063,11.10660171508789,12.059869581560836,17.767058684588687,13.517656020283676,0.8891457310949937,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_3/mix/HeladoNegro_MitadDelMundo - seg_3.wav
|
| 40 |
+
38,0.32333874702453613,0.6354581182207051,0.635458769332041,73.9389092628835,0.7380223161731874,9.420344352722168,11.304368533288283,15.631833304273673,14.104110837213394,0.8710827659749965,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_4/mix/HeladoNegro_MitadDelMundo - seg_4.wav
|
| 41 |
+
39,0.32333874702453613,0.6354581182207051,0.635458769332041,73.9389092628835,0.7380223161731874,9.420344352722168,11.304368533095417,15.631833294150344,14.10411084035519,0.8710827664107744,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_5/mix/HeladoNegro_MitadDelMundo - seg_5.wav
|
| 42 |
+
40,0.8644871711730957,1.3401706079454807,1.3401712429866564,74.2082751787392,0.6905740643470549,12.433319091796875,13.090904276118847,19.939782172958083,14.156270018323806,0.7792281138861851,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_7/mix/HeladoNegro_MitadDelMundo - seg_7.wav
|
| 43 |
+
41,0.8644871711730957,1.3401706079454807,1.3401712429866564,74.2082751787392,0.6905740643470549,12.433319091796875,13.090904281974037,19.939782170332954,14.156270026569445,0.7792281173649886,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_8/mix/HeladoNegro_MitadDelMundo - seg_8.wav
|
| 44 |
+
42,0.029872536659240723,0.04446954504281542,0.044470308946257475,70.96609327574804,0.7224079808793855,12.235817909240723,12.65727175944703,20.39518099062878,13.51222051653295,0.8328881159838999,../data/MedleyVox/duet/LewisAndClarke_TheSilverSea/seg_1/mix/LewisAndClarke_TheSilverSea - seg_1.wav
|
| 45 |
+
43,-0.13575482368469238,-0.10376259021913148,-0.10376208599585901,72.58028640014264,0.7092037857881859,10.23970651626587,11.06267151366577,16.846586747556287,12.550105964029513,0.857941306505643,../data/MedleyVox/duet/LewisAndClarke_TheSilverSea/seg_3/mix/LewisAndClarke_TheSilverSea - seg_3.wav
|
| 46 |
+
44,-0.0575982928276062,0.009280979185668503,0.00928115170552346,77.15117717584333,0.5501942882613997,0.1269608736038208,1.580562609236899,4.408985775666897,6.2648468853040225,0.5044691985288421,../data/MedleyVox/duet/LizNelson_ImComingHome/seg_10/mix/LizNelson_ImComingHome - seg_10.wav
|
| 47 |
+
45,0.44653069972991943,0.9149495947187788,0.9149509981631512,68.54832265739401,0.603216313324505,12.088458061218262,13.900688263320308,17.597805238789547,16.651102479620032,0.7662557420102396,../data/MedleyVox/duet/LizNelson_ImComingHome/seg_18/mix/LizNelson_ImComingHome - seg_18.wav
|
| 48 |
+
46,-0.07498586177825928,0.039093378896542186,0.03909472260584668,68.18195608026784,0.5613250548434863,12.173458576202393,13.217345631904005,18.644173283476697,14.817644739403507,0.7605089661983454,../data/MedleyVox/duet/LizNelson_Rainfall/seg_1/mix/LizNelson_Rainfall - seg_1.wav
|
| 49 |
+
47,-0.10924708843231201,-0.025565961708565732,-0.025564262496074797,67.47687677416485,0.5419227153844538,17.814453125,18.24319922037168,28.133332018859264,18.7268339804272,0.7422712172410768,../data/MedleyVox/duet/LizNelson_Rainfall/seg_4/mix/LizNelson_Rainfall - seg_4.wav
|
| 50 |
+
48,0.03569233417510986,0.07865928354784663,0.07866227152458255,64.71690736143401,0.7784548088637935,14.382667064666748,15.542269564732525,20.184675346380036,17.430887129109983,0.9010203527494615,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_12/mix/Meaxic_TakeAStep - seg_12.wav
|
| 51 |
+
49,-0.030388951301574707,0.338382970886777,0.3383872393066646,63.3121700591942,0.6576311937327939,9.546616077423096,10.951050868195512,14.88228654789271,13.400887420826034,0.8394029837060817,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_13/mix/Meaxic_TakeAStep - seg_13.wav
|
| 52 |
+
50,0.03510040044784546,0.10407529239339874,0.10407647943173759,68.82063189028952,0.7229356078319211,10.710720539093018,11.832052031532628,16.16570460989288,13.945462446838368,0.8334721824651311,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_16/mix/Meaxic_TakeAStep - seg_16.wav
|
| 53 |
+
51,0.03366875648498535,0.06401512478639715,0.06401668639898528,68.6193536138971,0.6888719068455467,11.985342502593994,12.957528434363267,18.259633696783176,14.658515054385926,0.8227090856333827,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_2/mix/Meaxic_TakeAStep - seg_2.wav
|
| 54 |
+
52,-0.015469789505004883,0.04791411982394389,0.0479163070336327,66.04103052168301,0.7195884365638618,10.362186908721924,11.441810501548073,16.162687012986446,13.382885445792372,0.8764368614356739,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_20/mix/Meaxic_TakeAStep - seg_20.wav
|
| 55 |
+
53,0.02606821060180664,0.05337666962720812,0.05337786920538717,68.62500613200521,0.6399336199827053,14.490634441375732,15.569125434865724,20.55151511581471,17.362489553737,0.8210313623856466,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_21/mix/Meaxic_TakeAStep - seg_21.wav
|
| 56 |
+
54,0.01994621753692627,0.08283956574448115,0.0828436623205393,63.764745710138925,0.7716457437549175,5.3831775188446045,8.668679697962054,12.931424127006398,13.248640766651347,0.8794507799240784,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_24/mix/Meaxic_TakeAStep - seg_24.wav
|
| 57 |
+
55,-0.3210386633872986,0.1289257082638503,0.12893011032451235,63.064773750362995,0.6766728957920127,8.020972728729248,9.398787207599316,13.441294635837448,11.789439476944462,0.8492208485796093,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_26/mix/Meaxic_TakeAStep - seg_26.wav
|
| 58 |
+
56,-0.08381712436676025,0.009783424745027358,0.009785366328946332,66.6573637694379,0.5412817772399335,8.232938766479492,9.693097707170882,12.216645573216304,13.509327399390425,0.8141078889277468,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_27/mix/Meaxic_TakeAStep - seg_27.wav
|
| 59 |
+
57,-0.00984618067741394,0.09944511890699925,0.0994473412910345,66.01563641914255,0.7121265309806494,9.024041175842285,10.204387836641096,14.364563255167345,12.528342029223117,0.847588594888092,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_3/mix/Meaxic_TakeAStep - seg_3.wav
|
| 60 |
+
58,0.004534482955932617,0.015004950507417458,0.015006999807690269,66.42676534256873,0.6288464557959668,14.446825504302979,15.339857055964256,21.341251884343713,16.631603186691265,0.8472144432243796,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_6/mix/Meaxic_TakeAStep - seg_6.wav
|
| 61 |
+
59,0.01832139492034912,0.040615419208266834,0.040619297814876276,63.593452124334604,0.7671799236989478,15.573838233947754,16.653474145087593,22.26576016547791,18.161218272155622,0.8864983748852379,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_7/mix/Meaxic_TakeAStep - seg_7.wav
|
| 62 |
+
60,0.024540424346923828,0.03968920467781922,0.039691168712646,67.78833739436185,0.7129239416590987,8.619178295135498,9.069582625108675,16.08357733942218,10.351933830942762,0.8169566151448673,../data/MedleyVox/duet/MidnightBlue_HuntingSeason/seg_1/mix/MidnightBlue_HuntingSeason - seg_1.wav
|
| 63 |
+
61,0.06482315063476562,0.08829956233767344,0.08830093059434718,69.2252761124445,0.6803583367215722,8.947620868682861,9.431076716877401,16.060427569190313,10.62509961644022,0.7894087624420311,../data/MedleyVox/duet/MidnightBlue_HuntingSeason/seg_2/mix/MidnightBlue_HuntingSeason - seg_2.wav
|
| 64 |
+
62,-0.053450942039489746,-0.02537573685695116,-0.025375229722147896,73.09378004500114,0.6852165526446382,14.812488079071045,15.716960220263807,22.75139238585732,16.850084553711294,0.8396342621303841,../data/MedleyVox/duet/NightPanther_Fire/seg_25/mix/NightPanther_Fire - seg_25.wav
|
| 65 |
+
63,-0.053450942039489746,-0.02537573685695116,-0.025375229722147896,73.09378004500114,0.6852165526446382,14.812488079071045,15.71696022391291,22.751392388110787,16.850084558050092,0.8396342652231448,../data/MedleyVox/duet/NightPanther_Fire/seg_26/mix/NightPanther_Fire - seg_26.wav
|
| 66 |
+
64,0.015639066696166992,0.06049708708367163,0.06049755399170342,73.84081477555415,0.7123269293927417,15.193305015563965,15.981418609131486,23.553648035843196,16.945183069333147,0.8471597684080632,../data/MedleyVox/duet/NightPanther_Fire/seg_27/mix/NightPanther_Fire - seg_27.wav
|
| 67 |
+
65,-0.08347582817077637,-0.045053952497162264,-0.045053524335429085,74.06878541658978,0.7024769612984867,14.817256450653076,15.47635782783294,23.16425263417217,16.378460746665382,0.8421815520083091,../data/MedleyVox/duet/NightPanther_Fire/seg_28/mix/NightPanther_Fire - seg_28.wav
|
| 68 |
+
66,-0.08347582817077637,-0.045053952497162264,-0.045053524335429085,74.06878541658978,0.7024769612984867,14.817256450653076,15.476357832913411,23.1642526386171,16.378460751279334,0.8421815502818765,../data/MedleyVox/duet/NightPanther_Fire/seg_29/mix/NightPanther_Fire - seg_29.wav
|
| 69 |
+
67,-0.05913543701171875,-0.0013635760030044075,-0.001363162109210947,74.31505052923924,0.725443566248053,15.611247539520264,16.554104174285328,23.580215903650906,17.658131053766997,0.8571602138943537,../data/MedleyVox/duet/NightPanther_Fire/seg_30/mix/NightPanther_Fire - seg_30.wav
|
| 70 |
+
68,-0.00010466575622558594,0.007721313286235265,0.00772217941793496,71.52947692191718,0.6337853923983667,17.565217971801758,18.32386034255247,25.489296872550717,19.34774745809819,0.8271004103500403,../data/MedleyVox/duet/NightPanther_Fire/seg_4/mix/NightPanther_Fire - seg_4.wav
|
| 71 |
+
69,0.03580141067504883,0.07034562586090809,0.07034649587253305,71.28746297341488,0.6933058934285098,19.837112426757812,20.627236851230883,27.999235543682264,21.67639603500656,0.8188435039226596,../data/MedleyVox/duet/NightPanther_Fire/seg_9/mix/NightPanther_Fire - seg_9.wav
|
| 72 |
+
70,-0.004528164863586426,-0.0006752473400262193,-0.0006751257847970038,78.87777029743508,0.6484113814514834,22.547690391540527,23.83109401876021,31.543988650117008,25.43049155670838,0.7354453925585309,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_1/mix/PatternIsMovement_SaveMe - seg_1.wav
|
| 73 |
+
71,0.006454944610595703,0.027474102508258547,0.027474720116547502,74.77703056436164,0.6379587910210298,7.817236065864563,9.041670754709155,13.333463783757324,11.402244683305357,0.6442025990669153,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_10/mix/PatternIsMovement_SaveMe - seg_10.wav
|
| 74 |
+
72,-0.007012367248535156,0.5283082348517265,0.5283091033306544,76.04506441339156,0.566961025132654,13.893479585647583,14.63928528340663,23.879904761943795,15.76190687042483,0.61834844261425,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_11/mix/PatternIsMovement_SaveMe - seg_11.wav
|
| 75 |
+
73,-0.034182071685791016,-0.016692467294078916,-0.01669150928697416,73.6993802691289,0.6475824584364226,12.920788764953613,15.065719155093538,22.931195283868032,17.45607241932059,0.6386827003742124,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_3/mix/PatternIsMovement_SaveMe - seg_3.wav
|
| 76 |
+
74,-0.044426918029785156,-0.02820762674483568,-0.028206713639735703,74.17369186093168,0.6114951574594413,9.953858852386475,10.489210986476728,18.232344878682508,11.401106699363085,0.6514953600708864,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_4/mix/PatternIsMovement_SaveMe - seg_4.wav
|
| 77 |
+
75,0.24094486236572266,0.3248105054102064,0.32481184826302467,74.13906328239187,0.6114891951797844,4.892396569252014,5.5901256744075996,12.541408577926068,7.090888711381266,0.5972937042327813,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_5/mix/PatternIsMovement_SaveMe - seg_5.wav
|
| 78 |
+
76,-0.06660699844360352,-0.04233689503281646,-0.04233602202458009,74.68749138809798,0.624064482033056,6.495960585772991,9.586392950221835,18.280591555440928,13.339349493192387,0.5649479136458724,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_6/mix/PatternIsMovement_SaveMe - seg_6.wav
|
| 79 |
+
77,0.06517410278320312,0.09466229169582174,0.0946630546188274,75.3102170569527,0.5814722964816406,6.222274631261826,9.447684039992476,18.102766973265986,13.348719644967172,0.5901038468051105,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_7/mix/PatternIsMovement_SaveMe - seg_7.wav
|
| 80 |
+
78,-2.384185791015625e-06,0.023244512062263745,0.023245223692976325,74.13777601514569,0.5891784481557799,14.312613487243652,15.477457081485431,19.123445085459135,18.129769742012613,0.6464204268921803,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_8/mix/PatternIsMovement_SaveMe - seg_8.wav
|
| 81 |
+
79,-2.384185791015625e-06,0.023244512062263745,0.023245223692976325,74.13777601514569,0.5891784481557799,14.312613487243652,15.477457081422173,19.123445088396835,18.12976973441171,0.6464204606795865,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_9/mix/PatternIsMovement_SaveMe - seg_9.wav
|
| 82 |
+
80,0.06983131170272827,0.2262668299605204,0.22626779140347425,69.71272862466647,0.6728658785161645,13.351724624633789,14.063441250454717,20.977630204420898,15.197026582904519,0.8681743523548715,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_10/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_10.wav
|
| 83 |
+
81,-0.08810663223266602,0.024716771635028945,0.024718017915818402,68.64311212645161,0.7049203037906818,12.461361408233643,13.422302079586075,19.714777105739234,14.972904742357589,0.8606360881532178,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_11/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_11.wav
|
| 84 |
+
82,-0.09375394880771637,0.0032477738240290766,0.003248607614388721,70.18527142287988,0.6928061219365298,12.723314762115479,13.109508907749905,21.962620653660323,13.767146482138658,0.873650288044461,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_12/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_12.wav
|
| 85 |
+
83,-0.014060363173484802,0.11516044901701089,0.11516153486062863,69.0975674303042,0.7189855998236807,12.172791481018066,12.778394885600012,20.324782132810917,13.666198812384867,0.8932758341044336,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_13/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_13.wav
|
| 86 |
+
84,-0.2536475658416748,-0.11520315724046826,-0.11520242109595535,70.714425500897,0.696504480636637,10.330602645874023,11.216682082904802,17.12326432248594,12.603815558476889,0.8823386277382355,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_14/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_14.wav
|
| 87 |
+
85,-0.005918450653553009,0.06893977914316507,0.06894075702605333,69.52297107055311,0.6902390457823746,11.499655723571777,12.217867297507098,19.033789654134367,13.329605806152635,0.8665538449667494,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_15/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_15.wav
|
| 88 |
+
86,-0.03271307051181793,0.08536172443577411,0.08536265179485872,69.7606227428021,0.6986759882493379,13.003861904144287,13.89870754879307,19.99864312898621,15.361186553757165,0.8762546522242805,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_16/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_16.wav
|
| 89 |
+
87,0.41347557306289673,0.5150859319428409,0.5150871983871239,68.92521266851992,0.7006205553097231,13.717366695404053,14.539850003370669,21.288721999365876,15.61873916259482,0.8638334705355162,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_2/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_2.wav
|
| 90 |
+
88,0.05131030082702637,0.13732700735583903,0.13732832000202205,68.34689487979576,0.6664581255999285,12.693491458892822,13.449052071495736,20.038943704103364,14.597287320619621,0.8350058445190043,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_3/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_3.wav
|
| 91 |
+
89,0.2281707525253296,0.35719551483669365,0.3571971712698294,67.48085035839381,0.6908935402739147,12.574252128601074,13.499704453675477,20.58499451735405,14.656920901844217,0.8758828946176651,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_4/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_4.wav
|
| 92 |
+
90,0.0011872053146362305,0.09006981289496263,0.09007151513415113,67.39091439792236,0.6627405721680095,13.81828498840332,14.693296633735407,21.381281539129947,15.88893772459354,0.8525871801924538,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_5/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_5.wav
|
| 93 |
+
91,-0.2059764862060547,-0.0783646740808206,-0.07836198814132889,66.56623278641678,0.6341275510310215,12.956748962402344,14.456798863264698,21.33465369263564,16.381887254438446,0.7845121327562494,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_6/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_6.wav
|
| 94 |
+
92,0.1308954954147339,0.18746186581107493,0.18746350745404938,67.97697580130765,0.6327386670824848,13.28718614578247,14.962441891936887,21.870981651049245,17.07745025421032,0.8007359464028635,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_7/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_7.wav
|
| 95 |
+
93,0.008746743202209473,0.1204589495075637,0.12045988244326716,69.7809053936839,0.6619015371233141,11.596556663513184,12.032876748012857,20.136484094378787,12.806190543429839,0.8406314952776188,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_9/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_9.wav
|
| 96 |
+
94,-0.12418636679649353,0.7830315363959708,0.7830321941163374,71.69840141488814,0.6662020488997313,-1.9109723344445229,0.2779284738139178,3.662535844847591,7.553217549239006,0.582640759669579,../data/MedleyVox/duet/TheTonTons_Lush/seg_2/mix/TheTonTons_Lush - seg_2.wav
|
| 97 |
+
95,0.24851877987384796,0.8722458150369499,0.8722464709581832,71.68481882161782,0.6805240299141901,2.9165565967559814,3.8661648528569437,6.025667829251082,8.932776184566503,0.6474665915443671,../data/MedleyVox/duet/TheTonTons_Lush/seg_8/mix/TheTonTons_Lush - seg_8.wav
|
| 98 |
+
96,0.3246358036994934,0.8879133404213468,0.8879140453987941,71.51922545990314,0.7588764593674855,7.648661375045776,9.61061494549563,12.886811664393512,12.598664472752802,0.8083985142473826,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_1/mix/TleilaxEnsemble_Late - seg_1.wav
|
| 99 |
+
97,0.05191130191087723,0.661137975861301,0.6611391591448432,69.00152197762479,0.7281064837455327,7.559250831604004,9.259028770546909,14.131196490086477,11.184348761526472,0.7592622017309785,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_2/mix/TleilaxEnsemble_Late - seg_2.wav
|
| 100 |
+
98,0.08045989274978638,0.5889291700501909,0.5889307228498019,67.94574504848656,0.7892001593364355,7.108006000518799,9.15906859991572,13.240735409011544,11.834526192730372,0.8004194034513405,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_3/mix/TleilaxEnsemble_Late - seg_3.wav
|
| 101 |
+
99,0.40290510654449463,1.6125075849580246,1.6125094111653124,68.20129283147544,0.5583285077738228,13.19340181350708,15.639135285246606,21.9346665113194,17.08064244462368,0.6103677758736703,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_1/mix/TleilaxEnsemble_MelancholyFlowers - seg_1.wav
|
| 102 |
+
100,0.49809932708740234,0.7600991654704372,0.7600994410362394,77.33930905820625,0.6401730764353525,13.34933090209961,15.17074234327722,19.596368137147685,17.235415811285158,0.6765862209734238,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_2/mix/TleilaxEnsemble_MelancholyFlowers - seg_2.wav
|
| 103 |
+
101,0.8220999240875244,3.638773063238361,3.6387732786596643,79.34839432599748,0.6701785766150965,13.877574443817139,15.796171133950908,20.979358314612377,17.451664256093352,0.6615754608118943,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_3/mix/TleilaxEnsemble_MelancholyFlowers - seg_3.wav
|
| 104 |
+
102,0.3838362693786621,0.8984425717933757,0.8984449612206196,66.18656648619641,0.6114197163557555,9.394717693328857,11.39471212542401,17.999787390238843,12.937138704886312,0.6095059643739164,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_4/mix/TleilaxEnsemble_MelancholyFlowers - seg_4.wav
|
| 105 |
+
103,0.3436286449432373,0.7613980301669554,0.7613991022064677,70.48092841517332,0.5994616128805523,18.75422954559326,19.757543724956175,26.615205536781986,20.77126128118131,0.6953689248195841,../data/MedleyVox/duet/Torres_NewSkin/seg_24/mix/Torres_NewSkin - seg_24.wav
|
| 106 |
+
104,0.09489607810974121,0.2741327260178008,0.2741336538010781,70.96130184486671,0.7301774734761826,21.627059936523438,23.5219300021381,29.890074819826765,24.921168405027974,0.8350600584942204,../data/MedleyVox/duet/Torres_NewSkin/seg_25/mix/Torres_NewSkin - seg_25.wav
|
| 107 |
+
105,0.016450166702270508,0.04560332256340405,0.04560368867644482,74.09980703566754,0.6848306897574168,19.58552074432373,20.91702168394515,26.81616725209708,22.414786262703082,0.8681859423489272,../data/MedleyVox/duet/Torres_NewSkin/seg_26/mix/Torres_NewSkin - seg_26.wav
|
| 108 |
+
106,0.025162339210510254,0.16673675310771952,0.1667372140825376,73.37675536639856,0.6494059456962972,21.012703895568848,22.080624913716015,27.836381424615254,23.496912154630074,0.8613912756463824,../data/MedleyVox/duet/Torres_NewSkin/seg_27/mix/Torres_NewSkin - seg_27.wav
|
| 109 |
+
107,0.010194927453994751,0.07056297637183873,0.07056333579826157,73.90828527169529,0.620881899555876,17.17660617828369,17.867931285567614,25.77149910558276,18.770522325072477,0.7566895832709519,../data/MedleyVox/duet/Torres_NewSkin/seg_28/mix/Torres_NewSkin - seg_28.wav
|
| 110 |
+
108,-0.022196829319000244,-0.011838552455158469,-0.011838197303164932,73.90252167903989,0.6757435707496644,16.013952255249023,16.549284015851946,24.556465923727835,17.320905816767024,0.8708037308966968,../data/MedleyVox/duet/Torres_NewSkin/seg_33/mix/Torres_NewSkin - seg_33.wav
|
| 111 |
+
109,0.0031448006629943848,0.0529103819973582,0.05291075528513278,73.77439123178598,0.677586277458446,14.60400390625,15.688950146468493,21.667799000403736,17.059799760150685,0.8669124810404665,../data/MedleyVox/duet/Torres_NewSkin/seg_34/mix/Torres_NewSkin - seg_34.wav
|
| 112 |
+
110,-0.041489750146865845,-0.015322006165329138,-0.015321705460128843,74.61934468959598,0.6555621176587945,14.184435367584229,15.133686065616242,20.9460200441965,16.492960843543777,0.8797259713550559,../data/MedleyVox/duet/Torres_NewSkin/seg_35/mix/Torres_NewSkin - seg_35.wav
|
| 113 |
+
111,0.04151415824890137,0.06827215647066587,0.06827252248999827,74.19974744323565,0.6804578262117269,12.162637710571289,13.19263271587976,18.285838908884102,15.058774892684497,0.864910567584005,../data/MedleyVox/duet/Torres_NewSkin/seg_36/mix/Torres_NewSkin - seg_36.wav
|
| 114 |
+
112,0.01772475242614746,0.04791080125620417,0.04791105665957884,75.64973622426785,0.646270524029401,9.376355648040771,10.484115282851723,14.537195669603175,12.894132259221667,0.8343454647914397,../data/MedleyVox/duet/Torres_NewSkin/seg_37/mix/Torres_NewSkin - seg_37.wav
|
| 115 |
+
113,0.019156694412231445,0.05764586077647649,0.05764618050284742,74.8098589143286,0.6841294832332929,10.968350887298584,11.569265054854775,18.087194272378138,12.741206240296963,0.8557209140884308,../data/MedleyVox/duet/Torres_NewSkin/seg_38/mix/Torres_NewSkin - seg_38.wav
|
| 116 |
+
114,-0.020918846130371094,0.020836558477576705,0.020836897101074836,75.00883776884778,0.6769045511385473,10.287030696868896,11.30102004343156,16.859231350681227,13.021486747996214,0.8437375900844306,../data/MedleyVox/duet/Torres_NewSkin/seg_39/mix/Torres_NewSkin - seg_39.wav
|
| 117 |
+
115,0.039122819900512695,0.10674367430600196,0.10674442504649573,73.15824295294688,0.6473504688872589,23.04123306274414,23.49933289906327,33.24709219061987,24.072161729112384,0.8040891845362853,../data/MedleyVox/duet/Torres_NewSkin/seg_45/mix/Torres_NewSkin - seg_45.wav
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.02406684309244156,
|
| 3 |
+
"input_sdr": 0.2720073709833361,
|
| 4 |
+
"input_sir": 0.2720112399176431,
|
| 5 |
+
"input_sar": 63.6579561715186,
|
| 6 |
+
"input_stoi": 0.8001933107257376,
|
| 7 |
+
"si_sdr": -0.020107148215174675,
|
| 8 |
+
"sdr": 2.3113227359507356,
|
| 9 |
+
"sir": 3.515217767392589,
|
| 10 |
+
"sar": 10.170321009767745,
|
| 11 |
+
"stoi": 0.7535998004890577,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/CatMartino_IPromise/seg_17/mix/CatMartino_IPromise - seg_17.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/mixture.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5a6dbab659e599d208d9d8f97c1a728125d3d152cba37d0137861c6cde160cd6
|
| 3 |
+
size 110060
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d40ad7c871cb9d87b1342f13805cae5d58a64c76938c50957f7431813fdb97ae
|
| 3 |
+
size 110060
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s0_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9792cc708f8df202ae84965581425650f97346cd811a86f71c2f0cf513489760
|
| 3 |
+
size 110060
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:152a9360b9941b428eee09b442a559d451e5aec823f276414de81bba1df66e86
|
| 3 |
+
size 110060
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_1/s1_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dd7508748819c4d13d408a5174338044e3f1276923a2e6fd42315218ed218430
|
| 3 |
+
size 110060
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": -0.02172267436981201,
|
| 3 |
+
"input_sdr": -0.01886554784865746,
|
| 4 |
+
"input_sir": -0.018864495288558936,
|
| 5 |
+
"input_sar": 69.34038650573328,
|
| 6 |
+
"input_stoi": 0.6960364677432993,
|
| 7 |
+
"si_sdr": 17.261977195739746,
|
| 8 |
+
"sdr": 17.636927828089505,
|
| 9 |
+
"sir": 27.249503047893555,
|
| 10 |
+
"sar": 18.154920651068643,
|
| 11 |
+
"stoi": 0.8144059119128467,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_4/mix/FilthyBird_IdLikeToKnow - seg_4.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/mixture.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d099b6861834fa46029943255ec302d795a415fad9a992921b24637997baa69
|
| 3 |
+
size 768046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de0f9dd71ef76a6bd7bb7988b43787c4a8d316f87c8bf32b25f65fdfccf31692
|
| 3 |
+
size 768046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s0_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c66d4bae4b52351237082fd367a4ef9f094f12be9880ef6c580106f26d39c691
|
| 3 |
+
size 768046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cdc46bf761f037e5e5bc476051e31e183485eac9000c577cafa0250ddfaa620e
|
| 3 |
+
size 768046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_28/s1_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1437b0bbf9ae9f041a4a122149d9b873bc9d3a51fe07c76ce48a8b5d9a7c7c9
|
| 3 |
+
size 768046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.0038509368896484375,
|
| 3 |
+
"input_sdr": 0.02030408099396208,
|
| 4 |
+
"input_sir": 0.02030560562826178,
|
| 5 |
+
"input_sar": 72.79871692374104,
|
| 6 |
+
"input_stoi": 0.5561395325418792,
|
| 7 |
+
"si_sdr": 15.383918285369873,
|
| 8 |
+
"sdr": 15.737125832356575,
|
| 9 |
+
"sir": 24.282791715101162,
|
| 10 |
+
"sar": 16.440832010953336,
|
| 11 |
+
"stoi": 0.6852088280754502,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/CatMartino_IPromise/seg_3/mix/CatMartino_IPromise - seg_3.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/mixture.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4de6d2df875730a33dab22b9d814fccdf6772c26edead69b580c9c8590a283e6
|
| 3 |
+
size 852046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:212940e85686b84cbb759124952e1b3732278a814da3433e20ec58595e39975d
|
| 3 |
+
size 852046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s0_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a451f2474716114e8acb8c647ea72debdc035006ccb9cb1396f1fbed4ff40124
|
| 3 |
+
size 852046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d6b2b4a856c19fe5545451f40f69ae90ef0e34f319c4319ee444f6489a230426
|
| 3 |
+
size 852046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_3/s1_estimate.wav
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d1b634c86ea3bf8431f43232fed7a926802c08993e68bb5a3e1e13f3d6e505c9
|
| 3 |
+
size 852046
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_34/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.02745962142944336,
|
| 3 |
+
"input_sdr": 0.09716946750039268,
|
| 4 |
+
"input_sir": 0.09717043462993269,
|
| 5 |
+
"input_sar": 71.83073233577244,
|
| 6 |
+
"input_stoi": 0.6162581878474078,
|
| 7 |
+
"si_sdr": 11.833811283111572,
|
| 8 |
+
"sdr": 12.116783113903214,
|
| 9 |
+
"sir": 21.407837997767192,
|
| 10 |
+
"sar": 12.761269858080432,
|
| 11 |
+
"stoi": 0.7401380382506599,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_18/mix/HeladoNegro_MitadDelMundo - seg_18.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_40/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.8644871711730957,
|
| 3 |
+
"input_sdr": 1.3401706079454807,
|
| 4 |
+
"input_sir": 1.3401712429866564,
|
| 5 |
+
"input_sar": 74.2082751787392,
|
| 6 |
+
"input_stoi": 0.6905740643470549,
|
| 7 |
+
"si_sdr": 12.433319091796875,
|
| 8 |
+
"sdr": 13.090904276118847,
|
| 9 |
+
"sir": 19.939782172958083,
|
| 10 |
+
"sar": 14.156270018323806,
|
| 11 |
+
"stoi": 0.7792281138861851,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_7/mix/HeladoNegro_MitadDelMundo - seg_7.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_61/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.06482315063476562,
|
| 3 |
+
"input_sdr": 0.08829956233767344,
|
| 4 |
+
"input_sir": 0.08830093059434718,
|
| 5 |
+
"input_sar": 69.2252761124445,
|
| 6 |
+
"input_stoi": 0.6803583367215722,
|
| 7 |
+
"si_sdr": 8.947620868682861,
|
| 8 |
+
"sdr": 9.431076716877401,
|
| 9 |
+
"sir": 16.060427569190313,
|
| 10 |
+
"sar": 10.62509961644022,
|
| 11 |
+
"stoi": 0.7894087624420311,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/MidnightBlue_HuntingSeason/seg_2/mix/MidnightBlue_HuntingSeason - seg_2.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_65/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": -0.08347582817077637,
|
| 3 |
+
"input_sdr": -0.045053952497162264,
|
| 4 |
+
"input_sir": -0.045053524335429085,
|
| 5 |
+
"input_sar": 74.06878541658978,
|
| 6 |
+
"input_stoi": 0.7024769612984867,
|
| 7 |
+
"si_sdr": 14.817256450653076,
|
| 8 |
+
"sdr": 15.47635782783294,
|
| 9 |
+
"sir": 23.16425263417217,
|
| 10 |
+
"sar": 16.378460746665382,
|
| 11 |
+
"stoi": 0.8421815520083091,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/NightPanther_Fire/seg_28/mix/NightPanther_Fire - seg_28.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_69/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.03580141067504883,
|
| 3 |
+
"input_sdr": 0.07034562586090809,
|
| 4 |
+
"input_sir": 0.07034649587253305,
|
| 5 |
+
"input_sar": 71.28746297341488,
|
| 6 |
+
"input_stoi": 0.6933058934285098,
|
| 7 |
+
"si_sdr": 19.837112426757812,
|
| 8 |
+
"sdr": 20.627236851230883,
|
| 9 |
+
"sir": 27.999235543682264,
|
| 10 |
+
"sar": 21.67639603500656,
|
| 11 |
+
"stoi": 0.8188435039226596,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/NightPanther_Fire/seg_9/mix/NightPanther_Fire - seg_9.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_72/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": -0.007012367248535156,
|
| 3 |
+
"input_sdr": 0.5283082348517265,
|
| 4 |
+
"input_sir": 0.5283091033306544,
|
| 5 |
+
"input_sar": 76.04506441339156,
|
| 6 |
+
"input_stoi": 0.566961025132654,
|
| 7 |
+
"si_sdr": 13.893479585647583,
|
| 8 |
+
"sdr": 14.63928528340663,
|
| 9 |
+
"sir": 23.879904761943795,
|
| 10 |
+
"sar": 15.76190687042483,
|
| 11 |
+
"stoi": 0.61834844261425,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_11/mix/PatternIsMovement_SaveMe - seg_11.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/examples/ex_79/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": -2.384185791015625e-06,
|
| 3 |
+
"input_sdr": 0.023244512062263745,
|
| 4 |
+
"input_sir": 0.023245223692976325,
|
| 5 |
+
"input_sar": 74.13777601514569,
|
| 6 |
+
"input_stoi": 0.5891784481557799,
|
| 7 |
+
"si_sdr": 14.312613487243652,
|
| 8 |
+
"sdr": 15.477457081422173,
|
| 9 |
+
"sir": 19.123445088396835,
|
| 10 |
+
"sar": 18.12976973441171,
|
| 11 |
+
"stoi": 0.6464204606795865,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_9/mix/PatternIsMovement_SaveMe - seg_9.wav"
|
| 13 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech/final_metrics.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"si_sdr": 13.008671734819252,
|
| 3 |
+
"si_sdr_imp": 12.979437920891133,
|
| 4 |
+
"sdr": 14.144418652672421,
|
| 5 |
+
"sdr_imp": 13.94410456184861,
|
| 6 |
+
"sir": 20.35606691634474,
|
| 7 |
+
"sir_imp": 20.155751768531648,
|
| 8 |
+
"sar": 15.937147104109295,
|
| 9 |
+
"sar_imp": -55.44671067284057,
|
| 10 |
+
"stoi": 0.787964090267542,
|
| 11 |
+
"stoi_imp": 0.12604789951048936
|
| 12 |
+
}
|
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/all_metrics.csv
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
,input_si_sdr,input_sdr,input_sir,input_sar,input_stoi,si_sdr,sdr,sir,sar,stoi,mix_path
|
| 2 |
+
0,0.0911402702331543,0.11333474572369884,0.11333570336043364,72.0603379655421,0.6399578763503541,15.402335166931152,16.130842314345358,23.430033266076975,17.139444905929338,0.8006582294841037,../data/MedleyVox/duet/CatMartino_IPromise/seg_1/mix/CatMartino_IPromise - seg_1.wav
|
| 3 |
+
1,0.02406684309244156,0.2720073709833361,0.2720112399176431,63.6579561715186,0.8001933107257376,0.13046014308929443,2.7290261784016705,4.242375912493816,9.642603157579616,0.7301137827089716,../data/MedleyVox/duet/CatMartino_IPromise/seg_17/mix/CatMartino_IPromise - seg_17.wav
|
| 4 |
+
2,0.023504503071308136,0.26802173592922623,0.2680257247729072,63.525063512012906,0.8064488134774055,0.1214735358953476,2.8136816344387974,4.198985947538819,10.018186977474704,0.7315904593726366,../data/MedleyVox/duet/CatMartino_IPromise/seg_21/mix/CatMartino_IPromise - seg_21.wav
|
| 5 |
+
3,0.0038509368896484375,0.02030408099396208,0.02030560562826178,72.79871692374104,0.5561395325418792,14.706861019134521,15.08914596032924,23.892912750640335,15.736805820599145,0.6883958670801427,../data/MedleyVox/duet/CatMartino_IPromise/seg_3/mix/CatMartino_IPromise - seg_3.wav
|
| 6 |
+
4,0.02143096923828125,0.07497634692502197,0.07497716754543271,73.59395962015613,0.5809720090023308,19.791841506958008,20.65748505193809,27.266692095313434,21.746619907955555,0.7652129637710159,../data/MedleyVox/duet/CatMartino_IPromise/seg_4/mix/CatMartino_IPromise - seg_4.wav
|
| 7 |
+
5,0.08760905265808105,0.13416892932315339,0.13416982711127678,70.56421253110653,0.8152177005156398,15.29309892654419,16.49203646672752,21.418083488229584,18.467843019219742,0.8541161867179938,../data/MedleyVox/duet/CatMartino_IPromise/seg_7/mix/CatMartino_IPromise - seg_7.wav
|
| 8 |
+
6,0.007455706596374512,0.018030733424175535,0.018031698569134134,69.55197961211707,0.7500427475771034,14.946728706359863,15.405367375333194,22.393261063077706,16.408024297209444,0.8875974928366954,../data/MedleyVox/duet/Cayetana_MissThing/seg_1/mix/Cayetana_MissThing - seg_1.wav
|
| 9 |
+
7,0.02685832977294922,0.03738885653335555,0.03738955054481585,71.38825034540149,0.6337365420795307,22.238048553466797,22.866352042196322,33.101131040450234,23.31095791410084,0.7507475322340527,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_45/mix/CelestialShore_DieForUs - seg_45.wav
|
| 10 |
+
8,0.05670976638793945,0.07074082917372415,0.07074144009968053,71.73390521848195,0.6185596781047933,21.2783260345459,21.919116631935417,32.173772715817414,22.383368733212706,0.7821663893841102,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_46/mix/CelestialShore_DieForUs - seg_46.wav
|
| 11 |
+
9,0.14332962036132812,0.16262581377419005,0.16262642313589504,71.86363741388934,0.6311706650878312,21.967731475830078,22.45804415189118,34.403509510108854,22.76861918220708,0.7367034342760046,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_47/mix/CelestialShore_DieForUs - seg_47.wav
|
| 12 |
+
10,0.128265380859375,0.15324910005827275,0.15324981439480334,71.23255351709889,0.6154861933701743,18.926352500915527,20.36774552181962,27.242030858198085,21.802312411712762,0.7498441392622395,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_48/mix/CelestialShore_DieForUs - seg_48.wav
|
| 13 |
+
11,0.05758732557296753,0.077363043101559,0.07736349149066901,72.95766862598383,0.6259839452526627,18.61725425720215,19.961480297621467,24.95831104130159,21.628910189476773,0.8339247660150169,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_49/mix/CelestialShore_DieForUs - seg_49.wav
|
| 14 |
+
12,0.019161254167556763,0.04505861820086385,0.04505906214744386,72.93893489031318,0.6886667461155986,16.35927963256836,18.124946854618287,22.08573354381185,20.692845240793048,0.8156329582725825,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_50/mix/CelestialShore_DieForUs - seg_50.wav
|
| 15 |
+
13,-0.017694473266601562,0.0046321481555779975,0.00463247629462793,74.25089108994348,0.635374454270375,17.09114646911621,18.052164549332694,24.504125488124025,19.204028357537368,0.8353317407029814,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_51/mix/CelestialShore_DieForUs - seg_51.wav
|
| 16 |
+
14,-0.00019288063049316406,0.005274560710137877,0.0052754997584760766,70.16690266362302,0.6232696854447997,20.2113094329834,20.744122818769984,30.283589359182805,21.260868437081292,0.7427037589750656,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_52/mix/CelestialShore_DieForUs - seg_52.wav
|
| 17 |
+
15,0.01592230796813965,0.025396623848236155,0.025397413560217963,70.82578262116812,0.6330517208512643,22.731372833251953,23.20323529683089,33.97650907576932,23.635536120687153,0.730406884941025,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_53/mix/CelestialShore_DieForUs - seg_53.wav
|
| 18 |
+
16,0.003073453903198242,0.033736624601256926,0.033737051234212556,73.53131209257162,0.6276588895157553,18.855237007141113,20.167683711859347,26.195236716700478,21.493381314922317,0.8064716388602231,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_54/mix/CelestialShore_DieForUs - seg_54.wav
|
| 19 |
+
17,0.011256307363510132,0.02321607777696777,0.0232165199284165,72.97352955149086,0.6509873487434956,16.427273750305176,17.95188560042172,23.10891156005569,19.78927172818502,0.8027545945255006,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_55/mix/CelestialShore_DieForUs - seg_55.wav
|
| 20 |
+
18,-0.20395523309707642,-0.17234465115083925,-0.17234428701259386,73.74059066668164,0.6297016626189389,17.009587287902832,17.74499494187654,25.446922094857094,18.597910447267232,0.7312610509241453,../data/MedleyVox/duet/CelestialShore_DieForUs/seg_56/mix/CelestialShore_DieForUs - seg_56.wav
|
| 21 |
+
19,-0.013024091720581055,0.04038188597472403,0.04038262248236868,71.23421055008231,0.6422089449986654,17.97057342529297,19.13103601922095,24.32440206314378,20.81638817101213,0.8677020027161323,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_1/mix/ClaraBerryAndWooldog_Boys - seg_1.wav
|
| 22 |
+
20,-0.0430682897567749,0.02098904110130828,0.020989503678424892,73.01358518576015,0.6004151029924082,14.168991565704346,16.18293916339372,20.800969032661335,19.118332286750327,0.8322823094368471,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_4/mix/ClaraBerryAndWooldog_Boys - seg_4.wav
|
| 23 |
+
21,0.005000650882720947,0.06638932945126408,0.06639005426505118,70.88366644865249,0.5542043543510237,13.198861598968506,13.992098214057151,20.21691704835004,15.30456776460132,0.7905641223484976,../data/MedleyVox/duet/ClaraBerryAndWooldog_Boys/seg_7/mix/ClaraBerryAndWooldog_Boys - seg_7.wav
|
| 24 |
+
22,0.09175297617912292,0.15464052461563546,0.15464089564175149,73.77624837319533,0.598772825417861,17.332879066467285,17.922089136777625,26.883004118899315,18.795282868601767,0.7881374295848629,../data/MedleyVox/duet/FacesOnFilm_WaitingForGa/seg_2/mix/FacesOnFilm_WaitingForGa - seg_2.wav
|
| 25 |
+
23,0.0015957951545715332,0.06626148109154995,0.06626181714834178,74.32902456467983,0.7016102840806384,15.836594581604004,16.77877844552136,20.861096782128847,19.001003160848647,0.834414403975626,../data/MedleyVox/duet/FamilyBand_Again/seg_1/mix/FamilyBand_Again - seg_1.wav
|
| 26 |
+
24,0.02108180522918701,0.07702722333964185,0.07702755382179804,74.34959429172716,0.6719157957177997,16.74653387069702,17.948988843252422,22.746818313128685,19.747700855102117,0.7963566741631255,../data/MedleyVox/duet/FamilyBand_Again/seg_2/mix/FamilyBand_Again - seg_2.wav
|
| 27 |
+
25,0.0210113525390625,0.44999612847186565,0.44999652144997726,74.61917157860216,0.5805111802982735,13.04970407485962,16.56113373466365,18.648567826729824,22.1451260114944,0.7357206409613395,../data/MedleyVox/duet/FamilyBand_Again/seg_6/mix/FamilyBand_Again - seg_6.wav
|
| 28 |
+
26,0.07987546920776367,0.13272754482617088,0.13272928396738415,69.83361102952001,0.6816841174395663,14.412473201751709,15.360268714883654,21.286205612277783,16.906826209894177,0.8347134859175794,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_1/mix/FilthyBird_IdLikeToKnow - seg_1.wav
|
| 29 |
+
27,0.004979491233825684,0.011056028115652039,0.011057087103157948,69.20021707551822,0.6649619360356808,21.04819965362549,21.568981578951902,30.393232364562436,22.18399211933879,0.8609591023739894,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_3/mix/FilthyBird_IdLikeToKnow - seg_3.wav
|
| 30 |
+
28,-0.02172267436981201,-0.01886554784865746,-0.018864495288558936,69.34038650573328,0.6960364677432993,18.713037490844727,18.97588887059632,28.903134305403043,19.47886794058659,0.8145691059232154,../data/MedleyVox/duet/FilthyBird_IdLikeToKnow/seg_4/mix/FilthyBird_IdLikeToKnow - seg_4.wav
|
| 31 |
+
29,-0.02795553207397461,0.19674289669623057,0.19674333759331475,74.84132036495171,0.6265701085920463,14.058848857879639,14.772434154145035,22.706287930569722,15.577675312231127,0.7200215088915922,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_11/mix/HeladoNegro_MitadDelMundo - seg_11.wav
|
| 32 |
+
30,-2.0922908782958984,-0.7448119301560405,-0.7448115538031477,75.3771459414981,0.6339736535992503,12.771053314208984,13.97239522650413,20.34045793530407,15.409365278524493,0.7857603194607587,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_13/mix/HeladoNegro_MitadDelMundo - seg_13.wav
|
| 33 |
+
31,-0.0640568733215332,0.18826454995281483,0.18826492101386982,74.51669068820232,0.6173298481221636,13.140581607818604,13.907966109807173,20.68875593377104,14.9748989215031,0.7296211765739998,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_14/mix/HeladoNegro_MitadDelMundo - seg_14.wav
|
| 34 |
+
32,0.031224727630615234,0.2817567609715672,0.2817577106486384,71.29556470748452,0.6797899387153046,12.017639636993408,13.101540830369439,18.3113477391354,14.7417651159469,0.8748444985241349,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_16/mix/HeladoNegro_MitadDelMundo - seg_16.wav
|
| 35 |
+
33,-0.29101765155792236,-0.13833882530884156,-0.13833836703988234,73.44498086768832,0.5691573747245798,11.861181735992432,12.686268294685192,18.812506714954527,14.168161985107004,0.7239317291007714,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_17/mix/HeladoNegro_MitadDelMundo - seg_17.wav
|
| 36 |
+
34,0.02745962142944336,0.09716946750039268,0.09717043462993269,71.83073233577244,0.6162581878474078,13.040217399597168,13.356428801878568,22.03852911198208,14.092058423030995,0.7340433346595794,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_18/mix/HeladoNegro_MitadDelMundo - seg_18.wav
|
| 37 |
+
35,-0.26619768142700195,-0.213581786451837,-0.21358096418494954,71.96222733630219,0.5950415002528179,12.805745124816895,13.165548833788613,21.796675697286915,14.069876164771337,0.7373993441094215,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_19/mix/HeladoNegro_MitadDelMundo - seg_19.wav
|
| 38 |
+
36,0.08208179473876953,0.1343153237494228,0.13431603118546098,72.45404927883251,0.5775131774705757,11.330626010894775,12.198156155636454,20.57013513654094,13.43920826720803,0.7149346482164269,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_20/mix/HeladoNegro_MitadDelMundo - seg_20.wav
|
| 39 |
+
37,-0.22986102104187012,0.029813558818715613,0.029814220321483287,72.73544176790732,0.7634997694258063,11.924163341522217,12.787518589128496,18.553824820999097,14.227838555748377,0.8959213903953267,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_3/mix/HeladoNegro_MitadDelMundo - seg_3.wav
|
| 40 |
+
38,0.32333874702453613,0.6354581182207051,0.635458769332041,73.9389092628835,0.7380223161731874,9.363210439682007,11.47344328399302,16.186858294335245,14.251760093949738,0.8843139272958443,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_4/mix/HeladoNegro_MitadDelMundo - seg_4.wav
|
| 41 |
+
39,0.32333874702453613,0.6354581182207051,0.635458769332041,73.9389092628835,0.7380223161731874,9.363215923309326,11.473452407864213,16.186892648567436,14.251754487024689,0.8843140887073457,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_5/mix/HeladoNegro_MitadDelMundo - seg_5.wav
|
| 42 |
+
40,0.8644871711730957,1.3401706079454807,1.3401712429866564,74.2082751787392,0.6905740643470549,13.680761337280273,14.287203856865954,20.8630187847365,15.414957722075652,0.8082144934445068,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_7/mix/HeladoNegro_MitadDelMundo - seg_7.wav
|
| 43 |
+
41,0.8644871711730957,1.3401706079454807,1.3401712429866564,74.2082751787392,0.6905740643470549,13.68073320388794,14.287194561906865,20.862992823865888,15.414953311924945,0.8082137370758771,../data/MedleyVox/duet/HeladoNegro_MitadDelMundo/seg_8/mix/HeladoNegro_MitadDelMundo - seg_8.wav
|
| 44 |
+
42,0.029872536659240723,0.04446954504281542,0.044470308946257475,70.96609327574804,0.7224079808793855,12.833311557769775,13.227689258173635,21.0839558987386,14.067089957767159,0.8231831387582333,../data/MedleyVox/duet/LewisAndClarke_TheSilverSea/seg_1/mix/LewisAndClarke_TheSilverSea - seg_1.wav
|
| 45 |
+
43,-0.13575482368469238,-0.10376259021913148,-0.10376208599585901,72.58028640014264,0.7092037857881859,10.410802364349365,11.14153755238628,17.176966373372302,12.513120792409874,0.8575159839447225,../data/MedleyVox/duet/LewisAndClarke_TheSilverSea/seg_3/mix/LewisAndClarke_TheSilverSea - seg_3.wav
|
| 46 |
+
44,-0.0575982928276062,0.009280979185668503,0.00928115170552346,77.15117717584333,0.5501942882613997,-0.5541253387928009,0.8684372682313628,3.6769479126431124,5.7153464725233505,0.49108742131289196,../data/MedleyVox/duet/LizNelson_ImComingHome/seg_10/mix/LizNelson_ImComingHome - seg_10.wav
|
| 47 |
+
45,0.44653069972991943,0.9149495947187788,0.9149509981631512,68.54832265739401,0.603216313324505,13.028499126434326,14.66569693718569,18.75934602598713,16.94270174144862,0.8023702600543556,../data/MedleyVox/duet/LizNelson_ImComingHome/seg_18/mix/LizNelson_ImComingHome - seg_18.wav
|
| 48 |
+
46,-0.07498586177825928,0.039093378896542186,0.03909472260584668,68.18195608026784,0.5613250548434863,11.534941673278809,12.511582122295524,17.783746436832637,14.192295408236586,0.7607495747725603,../data/MedleyVox/duet/LizNelson_Rainfall/seg_1/mix/LizNelson_Rainfall - seg_1.wav
|
| 49 |
+
47,-0.10924708843231201,-0.025565961708565732,-0.025564262496074797,67.47687677416485,0.5419227153844538,19.0550537109375,19.613485062139965,29.07604766573215,20.140871078349456,0.748374895872002,../data/MedleyVox/duet/LizNelson_Rainfall/seg_4/mix/LizNelson_Rainfall - seg_4.wav
|
| 50 |
+
48,0.03569233417510986,0.07865928354784663,0.07866227152458255,64.71690736143401,0.7784548088637935,11.978156089782715,13.4241312869529,16.878486250101226,16.60832554997834,0.8970073076593134,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_12/mix/Meaxic_TakeAStep - seg_12.wav
|
| 51 |
+
49,-0.030388951301574707,0.338382970886777,0.3383872393066646,63.3121700591942,0.6576311937327939,9.99910593032837,11.302395441829974,15.301548680827162,13.708203101552375,0.8545336701396211,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_13/mix/Meaxic_TakeAStep - seg_13.wav
|
| 52 |
+
50,0.03510040044784546,0.10407529239339874,0.10407647943173759,68.82063189028952,0.7229356078319211,11.033607482910156,11.959408762883218,16.626016653456396,13.887999716652258,0.8217486628817932,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_16/mix/Meaxic_TakeAStep - seg_16.wav
|
| 53 |
+
51,0.03366875648498535,0.06401512478639715,0.06401668639898528,68.6193536138971,0.6888719068455467,12.117576599121094,12.91976189806037,17.796142948373134,14.724543446565624,0.8199172472049256,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_2/mix/Meaxic_TakeAStep - seg_2.wav
|
| 54 |
+
52,-0.015469789505004883,0.04791411982394389,0.0479163070336327,66.04103052168301,0.7195884365638618,11.134056091308594,11.973735476087079,16.864839586940203,13.804036095605856,0.8954479360269672,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_20/mix/Meaxic_TakeAStep - seg_20.wav
|
| 55 |
+
53,0.02606821060180664,0.05337666962720812,0.05337786920538717,68.62500613200521,0.6399336199827053,15.04660177230835,15.871499277501078,21.618571542223933,17.385792509446432,0.8017585171266065,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_21/mix/Meaxic_TakeAStep - seg_21.wav
|
| 56 |
+
54,0.01994621753692627,0.08283956574448115,0.0828436623205393,63.764745710138925,0.7716457437549175,5.4520710706710815,8.702516085430268,13.403820286543986,13.218359740691358,0.8698656164233768,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_24/mix/Meaxic_TakeAStep - seg_24.wav
|
| 57 |
+
55,-0.3210386633872986,0.1289257082638503,0.12893011032451235,63.064773750362995,0.6766728957920127,8.8734769821167,10.117425869579751,14.280265931869081,12.42456068471529,0.8668960829261996,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_26/mix/Meaxic_TakeAStep - seg_26.wav
|
| 58 |
+
56,-0.08381712436676025,0.009783424745027358,0.009785366328946332,66.6573637694379,0.5412817772399335,8.813190937042236,9.928827888789208,12.903525732009928,13.258047490060257,0.845563487299025,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_27/mix/Meaxic_TakeAStep - seg_27.wav
|
| 59 |
+
57,-0.00984618067741394,0.09944511890699925,0.0994473412910345,66.01563641914255,0.7121265309806494,10.089964866638184,10.987593931524838,15.468189787992376,13.044932860491869,0.8634696594743718,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_3/mix/Meaxic_TakeAStep - seg_3.wav
|
| 60 |
+
58,0.004534482955932617,0.015004950507417458,0.015006999807690269,66.42676534256873,0.6288464557959668,15.427751541137695,16.120340946779603,22.848006242888424,17.215002725125647,0.8640129021547527,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_6/mix/Meaxic_TakeAStep - seg_6.wav
|
| 61 |
+
59,0.01832139492034912,0.040615419208266834,0.040619297814876276,63.593452124334604,0.7671799236989478,16.17048215866089,17.16636182142568,22.68654070046769,18.81825577850411,0.8879365554624066,../data/MedleyVox/duet/Meaxic_TakeAStep/seg_7/mix/Meaxic_TakeAStep - seg_7.wav
|
| 62 |
+
60,0.024540424346923828,0.03968920467781922,0.039691168712646,67.78833739436185,0.7129239416590987,8.410686492919922,8.778538488849165,15.099321410926219,10.15682379781965,0.8072139233564299,../data/MedleyVox/duet/MidnightBlue_HuntingSeason/seg_1/mix/MidnightBlue_HuntingSeason - seg_1.wav
|
| 63 |
+
61,0.06482315063476562,0.08829956233767344,0.08830093059434718,69.2252761124445,0.6803583367215722,9.304515600204468,9.678132114098947,16.273412617234964,10.86988939418352,0.7875149728326243,../data/MedleyVox/duet/MidnightBlue_HuntingSeason/seg_2/mix/MidnightBlue_HuntingSeason - seg_2.wav
|
| 64 |
+
62,-0.053450942039489746,-0.02537573685695116,-0.025375229722147896,73.09378004500114,0.6852165526446382,15.570566177368164,16.41018664553477,23.564858074880718,17.508022242733976,0.84973914823142,../data/MedleyVox/duet/NightPanther_Fire/seg_25/mix/NightPanther_Fire - seg_25.wav
|
| 65 |
+
63,-0.053450942039489746,-0.02537573685695116,-0.025375229722147896,73.09378004500114,0.6852165526446382,15.57056999206543,16.410200239674836,23.56486098713755,17.508039544565857,0.8497385153605052,../data/MedleyVox/duet/NightPanther_Fire/seg_26/mix/NightPanther_Fire - seg_26.wav
|
| 66 |
+
64,0.015639066696166992,0.06049708708367163,0.06049755399170342,73.84081477555415,0.7123269293927417,16.138899326324463,16.902997075434293,24.413265975051424,17.86795858292918,0.8591229974934024,../data/MedleyVox/duet/NightPanther_Fire/seg_27/mix/NightPanther_Fire - seg_27.wav
|
| 67 |
+
65,-0.08347582817077637,-0.045053952497162264,-0.045053524335429085,74.06878541658978,0.7024769612984867,15.631949424743652,16.233537285779676,23.446672891961192,17.220002836094213,0.8568082771619921,../data/MedleyVox/duet/NightPanther_Fire/seg_28/mix/NightPanther_Fire - seg_28.wav
|
| 68 |
+
66,-0.08347582817077637,-0.045053952497162264,-0.045053524335429085,74.06878541658978,0.7024769612984867,15.631958484649658,16.233549120121545,23.446685014409404,17.220016004381854,0.8568079537550863,../data/MedleyVox/duet/NightPanther_Fire/seg_29/mix/NightPanther_Fire - seg_29.wav
|
| 69 |
+
67,-0.05913543701171875,-0.0013635760030044075,-0.001363162109210947,74.31505052923924,0.725443566248053,16.510461807250977,17.385599625700152,24.62219368409633,18.42987124574762,0.8742893804447879,../data/MedleyVox/duet/NightPanther_Fire/seg_30/mix/NightPanther_Fire - seg_30.wav
|
| 70 |
+
68,-0.00010466575622558594,0.007721313286235265,0.00772217941793496,71.52947692191718,0.6337853923983667,18.744218826293945,19.36699719029776,26.922705870869606,20.300264757494077,0.8302519688078787,../data/MedleyVox/duet/NightPanther_Fire/seg_4/mix/NightPanther_Fire - seg_4.wav
|
| 71 |
+
69,0.03580141067504883,0.07034562586090809,0.07034649587253305,71.28746297341488,0.6933058934285098,20.87912368774414,21.5487670278686,31.34658389885465,22.247123632362097,0.8268329935949903,../data/MedleyVox/duet/NightPanther_Fire/seg_9/mix/NightPanther_Fire - seg_9.wav
|
| 72 |
+
70,-0.004528164863586426,-0.0006752473400262193,-0.0006751257847970038,78.87777029743508,0.6484113814514834,25.813542366027832,26.577393013521696,34.543043188822026,27.473052717041377,0.7520803829868787,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_1/mix/PatternIsMovement_SaveMe - seg_1.wav
|
| 73 |
+
71,0.006454944610595703,0.027474102508258547,0.027474720116547502,74.77703056436164,0.6379587910210298,9.369078397750854,10.244019682635756,15.782932674060701,12.128703123825964,0.6675554040979901,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_10/mix/PatternIsMovement_SaveMe - seg_10.wav
|
| 74 |
+
72,-0.007012367248535156,0.5283082348517265,0.5283091033306544,76.04506441339156,0.566961025132654,18.466877937316895,19.1106882670722,26.95048824464218,20.142246060132322,0.6432306471999574,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_11/mix/PatternIsMovement_SaveMe - seg_11.wav
|
| 75 |
+
73,-0.034182071685791016,-0.016692467294078916,-0.01669150928697416,73.6993802691289,0.6475824584364226,13.098896026611328,15.06460679072091,22.92572923661042,17.267630789605768,0.6932488291768364,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_3/mix/PatternIsMovement_SaveMe - seg_3.wav
|
| 76 |
+
74,-0.044426918029785156,-0.02820762674483568,-0.028206713639735703,74.17369186093168,0.6114951574594413,10.12113332748413,10.726376585549833,18.702794710149647,11.585110896385098,0.6357451047250965,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_4/mix/PatternIsMovement_SaveMe - seg_4.wav
|
| 77 |
+
75,0.24094486236572266,0.3248105054102064,0.32481184826302467,74.13906328239187,0.6114891951797844,8.45471978187561,9.29155454596246,15.151757714462686,10.909885117781451,0.6519332591632533,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_5/mix/PatternIsMovement_SaveMe - seg_5.wav
|
| 78 |
+
76,-0.06660699844360352,-0.04233689503281646,-0.04233602202458009,74.68749138809798,0.624064482033056,6.322635490447283,10.099009497238036,19.66343671097109,14.227570979613276,0.5906138776909984,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_6/mix/PatternIsMovement_SaveMe - seg_6.wav
|
| 79 |
+
77,0.06517410278320312,0.09466229169582174,0.0946630546188274,75.3102170569527,0.5814722964816406,6.632301680743694,9.755110605449278,18.428032474554136,13.555433295288672,0.6397924039153357,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_7/mix/PatternIsMovement_SaveMe - seg_7.wav
|
| 80 |
+
78,-2.384185791015625e-06,0.023244512062263745,0.023245223692976325,74.13777601514569,0.5891784481557799,17.31965923309326,18.504045692460664,25.576384104425394,20.100883732660677,0.6615945456890509,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_8/mix/PatternIsMovement_SaveMe - seg_8.wav
|
| 81 |
+
79,-2.384185791015625e-06,0.023244512062263745,0.023245223692976325,74.13777601514569,0.5891784481557799,17.319655895233154,18.5040332537914,25.57637748721089,20.100865692973056,0.6615934061040782,../data/MedleyVox/duet/PatternIsMovement_SaveMe/seg_9/mix/PatternIsMovement_SaveMe - seg_9.wav
|
| 82 |
+
80,0.06983131170272827,0.2262668299605204,0.22626779140347425,69.71272862466647,0.6728658785161645,13.757912635803223,14.484129280129125,21.301714312699186,15.680005877832237,0.883566039696026,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_10/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_10.wav
|
| 83 |
+
81,-0.08810663223266602,0.024716771635028945,0.024718017915818402,68.64311212645161,0.7049203037906818,12.734894275665283,13.568491876886812,20.13111445903788,15.069884541877473,0.8638152311705793,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_11/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_11.wav
|
| 84 |
+
82,-0.09375394880771637,0.0032477738240290766,0.003248607614388721,70.18527142287988,0.6928061219365298,13.517611026763916,13.84271967175484,22.831222792977307,14.512456479967447,0.8813885577076906,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_12/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_12.wav
|
| 85 |
+
83,-0.014060363173484802,0.11516044901701089,0.11516153486062863,69.0975674303042,0.7189855998236807,12.980441093444824,13.593661510314583,20.895355558085363,14.54638739396924,0.9000313997082057,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_13/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_13.wav
|
| 86 |
+
84,-0.2536475658416748,-0.11520315724046826,-0.11520242109595535,70.714425500897,0.696504480636637,11.134705066680908,12.024411963167326,17.78443520960832,13.48549990118569,0.899964906569227,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_14/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_14.wav
|
| 87 |
+
85,-0.005918450653553009,0.06893977914316507,0.06894075702605333,69.52297107055311,0.6902390457823746,11.86348009109497,12.494184383156387,19.303567936804633,13.630426390554941,0.8748915266507613,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_15/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_15.wav
|
| 88 |
+
86,-0.03271307051181793,0.08536172443577411,0.08536265179485872,69.7606227428021,0.6986759882493379,13.76018238067627,14.493708879358884,20.80465314340218,16.019859629642355,0.8859916374699606,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_16/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_16.wav
|
| 89 |
+
87,0.41347557306289673,0.5150859319428409,0.5150871983871239,68.92521266851992,0.7006205553097231,14.594369888305664,15.465646597027376,21.71125644938509,16.699993910723585,0.8681332515718458,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_2/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_2.wav
|
| 90 |
+
88,0.05131030082702637,0.13732700735583903,0.13732832000202205,68.34689487979576,0.6664581255999285,13.748909950256348,14.446126535168816,20.80187657977445,15.65719387179542,0.84904780776569,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_3/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_3.wav
|
| 91 |
+
89,0.2281707525253296,0.35719551483669365,0.3571971712698294,67.48085035839381,0.6908935402739147,12.936795711517334,13.860953392933144,21.149480938361982,15.111661262129156,0.8768993234747757,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_4/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_4.wav
|
| 92 |
+
90,0.0011872053146362305,0.09006981289496263,0.09007151513415113,67.39091439792236,0.6627405721680095,14.658803462982178,15.549913567663584,22.870117258185058,16.578064922635058,0.8552234299883132,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_5/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_5.wav
|
| 93 |
+
91,-0.2059764862060547,-0.0783646740808206,-0.07836198814132889,66.56623278641678,0.6341275510310215,13.25557804107666,14.755728701720194,21.67202250633325,16.83633852511109,0.7911126387071776,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_6/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_6.wav
|
| 94 |
+
92,0.1308954954147339,0.18746186581107493,0.18746350745404938,67.97697580130765,0.6327386670824848,13.179225444793701,14.957064377329235,22.04465155975057,17.23806232523109,0.8078375040711181,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_7/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_7.wav
|
| 95 |
+
93,0.008746743202209473,0.1204589495075637,0.12045988244326716,69.7809053936839,0.6619015371233141,12.62713623046875,13.033650733630445,20.722073502799585,13.881143910849381,0.8486356031382716,../data/MedleyVox/duet/PeterMatthewBauer_YouAlwaysLookForSomeoneLost/seg_9/mix/PeterMatthewBauer_YouAlwaysLookForSomeoneLost - seg_9.wav
|
| 96 |
+
94,-0.12418636679649353,0.7830315363959708,0.7830321941163374,71.69840141488814,0.6662020488997313,-2.090268611907959,0.39853817612483067,3.2390686583268335,8.209801835132986,0.5525797214820669,../data/MedleyVox/duet/TheTonTons_Lush/seg_2/mix/TheTonTons_Lush - seg_2.wav
|
| 97 |
+
95,0.24851877987384796,0.8722458150369499,0.8722464709581832,71.68481882161782,0.6805240299141901,2.1164785027503967,3.3045636147840933,5.362684113607389,8.715201821331629,0.6136667906444924,../data/MedleyVox/duet/TheTonTons_Lush/seg_8/mix/TheTonTons_Lush - seg_8.wav
|
| 98 |
+
96,0.3246358036994934,0.8879133404213468,0.8879140453987941,71.51922545990314,0.7588764593674855,6.318387985229492,8.084318767491599,11.088187050819606,11.549828494086904,0.7603937704988744,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_1/mix/TleilaxEnsemble_Late - seg_1.wav
|
| 99 |
+
97,0.05191130191087723,0.661137975861301,0.6611391591448432,69.00152197762479,0.7281064837455327,7.608973026275635,9.264703501277765,14.275425801932936,11.108024377743238,0.7185832671385586,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_2/mix/TleilaxEnsemble_Late - seg_2.wav
|
| 100 |
+
98,0.08045989274978638,0.5889291700501909,0.5889307228498019,67.94574504848656,0.7892001593364355,6.591590166091919,8.463996308465456,13.319248682726283,10.847107083008481,0.7726010479881793,../data/MedleyVox/duet/TleilaxEnsemble_Late/seg_3/mix/TleilaxEnsemble_Late - seg_3.wav
|
| 101 |
+
99,0.40290510654449463,1.6125075849580246,1.6125094111653124,68.20129283147544,0.5583285077738228,13.266005992889404,15.738650730546219,22.274681575509323,17.16781791655719,0.5915944417455665,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_1/mix/TleilaxEnsemble_MelancholyFlowers - seg_1.wav
|
| 102 |
+
100,0.49809932708740234,0.7600991654704372,0.7600994410362394,77.33930905820625,0.6401730764353525,13.751083850860596,15.695108266600664,20.506878108724482,17.563059014771476,0.6529775082785796,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_2/mix/TleilaxEnsemble_MelancholyFlowers - seg_2.wav
|
| 103 |
+
101,0.8220999240875244,3.638773063238361,3.6387732786596643,79.34839432599748,0.6701785766150965,15.48633861541748,17.974491807184013,22.85235555719465,19.719057900293777,0.6518516780572822,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_3/mix/TleilaxEnsemble_MelancholyFlowers - seg_3.wav
|
| 104 |
+
102,0.3838362693786621,0.8984425717933757,0.8984449612206196,66.18656648619641,0.6114197163557555,9.805960178375244,11.858537045269763,18.155749471426287,13.464967280849136,0.5900329994531615,../data/MedleyVox/duet/TleilaxEnsemble_MelancholyFlowers/seg_4/mix/TleilaxEnsemble_MelancholyFlowers - seg_4.wav
|
| 105 |
+
103,0.3436286449432373,0.7613980301669554,0.7613991022064677,70.48092841517332,0.5994616128805523,19.5114164352417,20.9335171172593,27.216765098684995,22.111205070080395,0.6648458535709365,../data/MedleyVox/duet/Torres_NewSkin/seg_24/mix/Torres_NewSkin - seg_24.wav
|
| 106 |
+
104,0.09489607810974121,0.2741327260178008,0.2741336538010781,70.96130184486671,0.7301774734761826,21.70561695098877,24.07687757640293,31.17668513497446,25.266738438183964,0.8375523829017242,../data/MedleyVox/duet/Torres_NewSkin/seg_25/mix/Torres_NewSkin - seg_25.wav
|
| 107 |
+
105,0.016450166702270508,0.04560332256340405,0.04560368867644482,74.09980703566754,0.6848306897574168,20.605926513671875,21.93825434730713,27.899479532207966,23.355614555765094,0.8803196585945201,../data/MedleyVox/duet/Torres_NewSkin/seg_26/mix/Torres_NewSkin - seg_26.wav
|
| 108 |
+
106,0.025162339210510254,0.16673675310771952,0.1667372140825376,73.37675536639856,0.6494059456962972,21.234837532043457,22.494310623628394,28.832564239790837,23.7616760975542,0.8743633676260401,../data/MedleyVox/duet/Torres_NewSkin/seg_27/mix/Torres_NewSkin - seg_27.wav
|
| 109 |
+
107,0.010194927453994751,0.07056297637183873,0.07056333579826157,73.90828527169529,0.620881899555876,20.415751457214355,21.40644035821233,28.21592355194172,22.444261013525228,0.7550193421742621,../data/MedleyVox/duet/Torres_NewSkin/seg_28/mix/Torres_NewSkin - seg_28.wav
|
| 110 |
+
108,-0.022196829319000244,-0.011838552455158469,-0.011838197303164932,73.90252167903989,0.6757435707496644,16.644835472106934,17.057778891313028,25.982898594001483,17.66398541788536,0.8817834714752422,../data/MedleyVox/duet/Torres_NewSkin/seg_33/mix/Torres_NewSkin - seg_33.wav
|
| 111 |
+
109,0.0031448006629943848,0.0529103819973582,0.05291075528513278,73.77439123178598,0.677586277458446,15.364448547363281,16.47298769688608,22.126071947415777,17.951900685440698,0.8820027908193199,../data/MedleyVox/duet/Torres_NewSkin/seg_34/mix/Torres_NewSkin - seg_34.wav
|
| 112 |
+
110,-0.041489750146865845,-0.015322006165329138,-0.015321705460128843,74.61934468959598,0.6555621176587945,15.16216516494751,15.984395704288271,21.94587509319878,17.28193077117986,0.8963885193357035,../data/MedleyVox/duet/Torres_NewSkin/seg_35/mix/Torres_NewSkin - seg_35.wav
|
| 113 |
+
111,0.04151415824890137,0.06827215647066587,0.06827252248999827,74.19974744323565,0.6804578262117269,12.455653190612793,13.480076714664412,19.557557878438736,15.203382421118219,0.8731367968480124,../data/MedleyVox/duet/Torres_NewSkin/seg_36/mix/Torres_NewSkin - seg_36.wav
|
| 114 |
+
112,0.01772475242614746,0.04791080125620417,0.04791105665957884,75.64973622426785,0.646270524029401,5.735335350036621,7.132180185571285,10.54227863349073,10.24154449407879,0.8198489251535843,../data/MedleyVox/duet/Torres_NewSkin/seg_37/mix/Torres_NewSkin - seg_37.wav
|
| 115 |
+
113,0.019156694412231445,0.05764586077647649,0.05764618050284742,74.8098589143286,0.6841294832332929,11.313482284545898,11.75018564882036,18.496191876023044,12.855568725144344,0.8678501633710554,../data/MedleyVox/duet/Torres_NewSkin/seg_38/mix/Torres_NewSkin - seg_38.wav
|
| 116 |
+
114,-0.020918846130371094,0.020836558477576705,0.020836897101074836,75.00883776884778,0.6769045511385473,10.813549518585205,11.777331552320074,17.618965302133894,13.624159433628265,0.8572373864935922,../data/MedleyVox/duet/Torres_NewSkin/seg_39/mix/Torres_NewSkin - seg_39.wav
|
| 117 |
+
115,0.039122819900512695,0.10674367430600196,0.10674442504649573,73.15824295294688,0.6473504688872589,23.593698501586914,24.210496224972324,35.85482808202137,24.609093815272026,0.7980296811331213,../data/MedleyVox/duet/Torres_NewSkin/seg_45/mix/Torres_NewSkin - seg_45.wav
|
MedleyVox-MultiSinger/eval_results/singing_librispeech_iSRNet/examples/ex_16/metrics.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"input_si_sdr": 0.003073453903198242,
|
| 3 |
+
"input_sdr": 0.033736624601256926,
|
| 4 |
+
"input_sir": 0.033737051234212556,
|
| 5 |
+
"input_sar": 73.53131209257162,
|
| 6 |
+
"input_stoi": 0.6276588895157553,
|
| 7 |
+
"si_sdr": 18.855237007141113,
|
| 8 |
+
"sdr": 20.167683711859347,
|
| 9 |
+
"sir": 26.195236716700478,
|
| 10 |
+
"sar": 21.493381314922317,
|
| 11 |
+
"stoi": 0.8064716388602231,
|
| 12 |
+
"mix_path": "../data/MedleyVox/duet/CelestialShore_DieForUs/seg_54/mix/CelestialShore_DieForUs - seg_54.wav"
|
| 13 |
+
}
|