Hugging Face's logo

lainlives
/
audio-separator-models-old

ONNX
Model card Files
xet
Community
lainlives commited on
Commit
8c06eaf
·
verified ·
1 Parent(s): d6bd41f

Add files using upload-large-folder tool

Browse files
Files changed (6) hide show
  1. bs_roformer_vocals_gabox.ckpt.aria2 +0 -0
  2. bs_roformer_vocals_revive_unwa.ckpt.aria2 +0 -0
  3. bs_roformer_vocals_revive_v2_unwa.ckpt.aria2 +0 -0
  4. bs_roformer_vocals_revive_v3e_unwa.ckpt.aria2 +0 -0
  5. config_dereverb_echo_mel_band_roformer_v2.yaml.1 +64 -0
  6. config_dereverb_echo_mel_band_roformer_v2.yaml.2 +64 -0
bs_roformer_vocals_gabox.ckpt.aria2 ADDED
Binary file (195 Bytes). View file
 
bs_roformer_vocals_revive_unwa.ckpt.aria2 ADDED
Binary file (215 Bytes). View file
 
bs_roformer_vocals_revive_v2_unwa.ckpt.aria2 ADDED
Binary file (215 Bytes). View file
 
bs_roformer_vocals_revive_v3e_unwa.ckpt.aria2 ADDED
Binary file (215 Bytes). View file
 
config_dereverb_echo_mel_band_roformer_v2.yaml.1 ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ audio:
2
+ chunk_size: 352800
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: 0.000
10
+
11
+ model:
12
+ dim: 256
13
+ depth: 8
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: 0.1
22
+ ff_dropout: 0.1
23
+ flash_attn: True
24
+ dim_freqs_in: 1025
25
+ sample_rate: 44100
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: 1
43
+ gradient_accumulation_steps: 8
44
+ grad_clip: 0
45
+ instruments:
46
+ - dry
47
+ - other
48
+ lr: 1.0e-05
49
+ patience: 2
50
+ reduce_factor: 0.95
51
+ target_instrument: dry
52
+ num_epochs: 1000
53
+ num_steps: 1000
54
+ q: 0.95
55
+ coarse_loss_clip: true
56
+ ema_momentum: 0.999
57
+ optimizer: adam
58
+ other_fix: false
59
+ use_amp: true
60
+
61
+ inference:
62
+ batch_size: 1
63
+ dim_t: 801
64
+ num_overlap: 4
config_dereverb_echo_mel_band_roformer_v2.yaml.2 ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ audio:
2
+ chunk_size: 352800
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: 0.000
10
+
11
+ model:
12
+ dim: 256
13
+ depth: 8
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: 0.1
22
+ ff_dropout: 0.1
23
+ flash_attn: True
24
+ dim_freqs_in: 1025
25
+ sample_rate: 44100
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: 1
43
+ gradient_accumulation_steps: 8
44
+ grad_clip: 0
45
+ instruments:
46
+ - dry
47
+ - other
48
+ lr: 1.0e-05
49
+ patience: 2
50
+ reduce_factor: 0.95
51
+ target_instrument: dry
52
+ num_epochs: 1000
53
+ num_steps: 1000
54
+ q: 0.95
55
+ coarse_loss_clip: true
56
+ ema_momentum: 0.999
57
+ optimizer: adam
58
+ other_fix: false
59
+ use_amp: true
60
+
61
+ inference:
62
+ batch_size: 1
63
+ dim_t: 801
64
+ num_overlap: 4