Add files using upload-large-folder tool

bsroformers/bs_karaoke_gabox_IS.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db8357825398d4231031ad1ab4aa12a94bcaad8d67e8ce5e4b3c5b48fdee1d4f
+size 204483448

bsroformers/karaoke_bs_roformer.yaml

@@ -0,0 +1,129 @@
+audio:
+  chunk_size: 352800
+  dim_f: 1024
+  dim_t: 801 # don't work (use in model)
+  hop_length: 441 # don't work (use in model)
+  n_fft: 2048
+  num_channels: 2
+  sample_rate: 44100
+  min_mean_abs: 0.000
+
+model:
+  dim: 256
+  depth: 12
+  stereo: true
+  num_stems: 1
+  time_transformer_depth: 1
+  freq_transformer_depth: 1
+  linear_transformer_depth: 0
+  freqs_per_bands: !!python/tuple
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 2
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 4
+    - 12
+    - 12
+    - 12
+    - 12
+    - 12
+    - 12
+    - 12
+    - 12
+    - 24
+    - 24
+    - 24
+    - 24
+    - 24
+    - 24
+    - 24
+    - 24
+    - 48
+    - 48
+    - 48
+    - 48
+    - 48
+    - 48
+    - 48
+    - 48
+    - 128
+    - 129
+  dim_head: 64
+  heads: 8
+  attn_dropout: 0.0
+  ff_dropout: 0.0
+  flash_attn: true
+  dim_freqs_in: 1025
+  stft_n_fft: 2048
+  stft_hop_length: 512
+  stft_win_length: 2048
+  stft_normalized: false
+  mask_estimator_depth: 2
+  multi_stft_resolution_loss_weight: 1.0
+  multi_stft_resolutions_window_sizes: !!python/tuple
+  - 4096
+  - 2048
+  - 1024
+  - 512
+  - 256
+  multi_stft_hop_size: 147
+  multi_stft_normalized: False
+  mlp_expansion_factor: 4
+  use_torch_checkpoint: True
+  skip_connection: False
+
+
+training:
+  batch_size: 1
+  gradient_accumulation_steps: 999
+  grad_clip: 1
+  instruments: ['vocals', 'other']
+  lr: 1.0e-5
+  patience: 1000000
+  reduce_factor: 0.75
+  target_instrument: vocals
+  num_epochs: 1000
+  num_steps: 1000
+  q: 0.95
+  coarse_loss_clip: true
+  ema_momentum: 0.999
+  optimizer: Fira
+  other_fix: True # it's needed for checking on multisong dataset if other is actually instrumental
+  use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
+  use_torch_checkpoint: True
+
+inference:
+  batch_size: 6
+  dim_t: 1251
+  num_overlap: 2

melbandroformers/experimental/BS_ResurrectioN.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d28b3a422531e1ca482d29302a5b67614eb45c09d72ce954258011ce24ef0919
+size 204483033

melbandroformers/experimental/Fullness.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c67f47fa84146c8182f0061c194a51782439fe732e6ffecb810f0e4af7026f6d
+size 913091027

melbandroformers/experimental/INSTV10.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6eb987c3fd93c66adfea2e4284d1da668e80a98f75b6436b4be454780a728182
+size 913026650

melbandroformers/experimental/INSTV8.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d51dae045064a5817cd7588c46bfe2f93542d4dbe66c1e1a6e75ca081299a53
+size 913026650

melbandroformers/experimental/INSTV8N.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25920c876e601d4ccbf1684b19ecdac4b9fcfcc7f48d2c095d81040ec3fecbea
+size 913026650

melbandroformers/experimental/INSTV9.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b689ffa942c48353bbad9e8df7b96d61a24b5c50256b0052579b327552565c8
+size 913026650

melbandroformers/experimental/Inst_FV8b.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a28f79ae38b92caaeb37c5ccdf3912316b135414353ec0d5553d507e3de8ef7
+size 913027055

melbandroformers/experimental/Inst_Fv8.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:026caebb240226f64b1611e623735bb14c042f772464292aa1a26e896a759cc5
+size 913027055

melbandroformers/experimental/Inst_Fv9.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6c02cd154bb49d6d08044a8dc935a94ec26d09f2a0e6a03cc80a65d05a98ff2
+size 913090472

melbandroformers/experimental/Karaoke_GaboxV2.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec34be50327aeaf1a996c27977f5c30d1ac80c0076d69683d3e5184c31ea29d3
+size 913090472

melbandroformers/experimental/Lead_VocalDereverb.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:097c13b298414905651867ef315d881bbf2604a5c04be28820ed5946338b3eb4
+size 913031195

melbandroformers/experimental/inst_fv7b.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3cf9538616c835575eb07bef1fe4550b2406c852919a5644d22e9ae84a2ec6ba
+size 913030778

melbandroformers/experimental/instv7beta.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b4f12d81fd7fb02f38609216d59f0e42b3dca655fd90ca275ab5321b3e4d9aa
+size 913026650

melbandroformers/experimental/instv7beta2.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11dcd18b2bca5ff189924145a52c9e1a145632c7659aac59b239221dca4f1703
+size 913026650

melbandroformers/experimental/instv7beta3.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc33cc26ab805b057e47a753241bc17102564af4bc1c0e76364f3bb6a31c09d2
+size 913026650

melbandroformers/experimental/instv7plus.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7baf1aab873e534a186e2dcf6e64bebad2dbf3e2c49337729487294539118232
+size 913031195

melbandroformers/experimental/kar_gabox.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:303fc631e7aa587e9dc1e6ac4bb3667c6ba53aacb6b6a90abcfcf57935b92bd8
+size 913026650

melbandroformers/experimental/small_inst.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:31c0a91d60aac2bdc6ee4a53bf7ebd0370e6e894f6b22a1e37511e05eecca335
+size 202573672

melbandroformers/experimental/vocfv7beta1.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6bb98488b80e7797e6f447c0dd8f89f2b7040eebc5e839b5a59871b7a3a19265
+size 913091027

melbandroformers/experimental/vocfv7beta2.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85bcaa03568f96e56b245302c177658e8cb83aa6e2f4ae260e67202db9d56123
+size 913030778

melbandroformers/experimental/vocfv7beta3.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45463b0051651b7ed15677cc41f21bd384d66796938f46955617bf6f1d661988
+size 913030778

melbandroformers/instrumental/INSTV5.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38e935cf1e97afcc1de84d0bdb87dd8090bad530fa0df28e707d16448e1d38e2
+size 913026650

melbandroformers/instrumental/INSTV5N.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:175203923fac3e52ae00e7e37d41e8a7fef5020b6ee4e4144f4786daabc54b34
+size 913026650

melbandroformers/instrumental/INSTV6.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:677951b8556a27abe32e39705640638826e78101fa901a51ad73d20522be6d25
+size 913026650

melbandroformers/instrumental/INSTV6N.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:802f3e5d183d7c4b50dea147c320e61634f5be6ff55fa899fdebeaf0f3cf7f42
+size 913026650

melbandroformers/instrumental/INSTV7N.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0ca36af5d1314be46b56c8a53b6be02f98511fa5d7e3e196fd895755e65be3c
+size 913026650

melbandroformers/instrumental/Inst_ExperimentalV1.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb19b968287d31b6d229816342aa7f08f7e4010b7c61be2427fbdb0c04a8a020
+size 913026650

melbandroformers/instrumental/Inst_GaboxFVX.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:545ef13b0cdbac505818a38db98e09c54e7c03ea17b4e0c895a531bfa352fa59
+size 913026650

melbandroformers/instrumental/Inst_GaboxFv7z.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef229f0dcd370c1767e4292981c59e5248814da45f32bfacebcc0f28adaa30a1
+size 913031195

melbandroformers/instrumental/Inst_GaboxFv8.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50bfa127d21f419e0da89730867d28c7ac4484c9473e6f313d036bc8b429df80
+size 913026650

melbandroformers/instrumental/Inst_GaboxFv9.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:068d29a54e71f69ed871ca410e09b9877e153ad1439e825bd093573ea4de5762
+size 913091027

melbandroformers/instrumental/Inst_GaboxV7.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e725a860176acb475d983a1ddd9c1a99a619c69cc9ceda808dd294d10db746a5
+size 913026650

melbandroformers/instrumental/denoisedebleed.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91aa7a546ed2e93482e4629c982d35b0d258bb3de6eeab497fd91658cc86c7fd
+size 913026650

melbandroformers/instrumental/inst_Fv4.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b03d9bf0fc643dabf006dee38fd798140b4884c1b65c573b4488c1a2a876ad84
+size 913091027

melbandroformers/instrumental/inst_Fv4Noise.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f50296e913b9af3b5b3b961e92877ef0d4a74f9a433e796e89960c4c2b1abe53
+size 913026650

melbandroformers/instrumental/inst_gabox.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b3ad6bd8bed3aaaa4d9320ea2ca910d140196a2302186db1754f3a8d8e16fb1
+size 913026650

melbandroformers/instrumental/inst_gabox.yaml

@@ -0,0 +1,51 @@
+audio:
+  chunk_size: 485100
+  dim_f: 1024
+  dim_t: 1101
+  hop_length: 441
+  n_fft: 2048
+  num_channels: 2
+  sample_rate: 44100
+  min_mean_abs: 0.000
+
+model:
+  dim: 384
+  depth: 6
+  stereo: true
+  num_stems: 1
+  time_transformer_depth: 1
+  freq_transformer_depth: 1
+  num_bands: 60
+  dim_head: 64
+  heads: 8
+  attn_dropout: 0
+  ff_dropout: 0
+  flash_attn: True
+  dim_freqs_in: 1025
+  sample_rate: 44100  # needed for mel filter bank from librosa
+  stft_n_fft: 2048
+  stft_hop_length: 441
+  stft_win_length: 2048
+  stft_normalized: False
+  mask_estimator_depth: 2
+  multi_stft_resolution_loss_weight: 1.0
+  multi_stft_resolutions_window_sizes: !!python/tuple
+  - 4096
+  - 2048
+  - 1024
+  - 512
+  - 256
+  multi_stft_hop_size: 147
+  multi_stft_normalized: False
+
+training:
+  instruments:
+  - Instrumental
+  - Vocals
+  target_instrument: Instrumental
+  use_amp: True
+
+inference:
+  batch_size: 1
+  dim_t: 1101
+  num_overlap: 2

melbandroformers/instrumental/inst_gaboxBv1.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de972fb724601beef237abe94c8b934c73218e9baf3e344ab4c2333276e5cfe7
+size 913026650

melbandroformers/instrumental/intrumental_gabox.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b3ad6bd8bed3aaaa4d9320ea2ca910d140196a2302186db1754f3a8d8e16fb1
+size 913026650

melbandroformers/instrumental/v10.yaml

@@ -0,0 +1,73 @@
+audio:
+  chunk_size: 352800
+  dim_f: 1024
+  dim_t: 256
+  hop_length: 441
+  n_fft: 2048
+  num_channels: 2
+  sample_rate: 44100
+  min_mean_abs: 0.00
+
+model:
+  dim: 256
+  depth: 12
+  stereo: true
+  num_stems: 1
+  time_transformer_depth: 1
+  freq_transformer_depth: 1
+  num_bands: 60
+  dim_head: 64
+  heads: 8
+  attn_dropout: 0
+  ff_dropout: 0
+  flash_attn: true
+  dim_freqs_in: 1025
+  sample_rate: 44100
+  stft_n_fft: 2048
+  stft_hop_length: 441
+  stft_win_length: 2048
+  stft_normalized: true
+  mask_estimator_depth: 2
+  multi_stft_resolution_loss_weight: 1.0
+  multi_stft_resolutions_window_sizes: !!python/tuple
+  - 4096
+  - 2048
+  - 1024
+  - 512
+  - 256
+  multi_stft_hop_size: 250
+  multi_stft_normalized: false
+  use_torch_checkpoint: true
+
+training:
+  batch_size: 1
+  gradient_accumulation_steps: 999999999999999999999999
+  grad_clip: 0
+  instruments:
+  - other
+  - vocals
+  lr: 0.00001
+  patience: 100000000
+  reduce_factor: 0.95
+  target_instrument: other
+  num_epochs: 1000
+  num_steps: 1000
+  augmentation: false # enable augmentations by audiomentations and pedalboard
+  augmentation_type: simple1
+  use_mp3_compress: false # Deprecated
+  augmentation_mix: false # Mix several stems of the same type with some probability
+  augmentation_loudness: true # randomly change loudness of each stem
+  augmentation_loudness_type: 1 # Type 1 or 2
+  augmentation_loudness_min: 0
+  augmentation_loudness_max: 0
+  q: 0.95
+  coarse_loss_clip: false
+  ema_momentum: 0.999
+  optimizer: adamw
+  other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
+  use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
+
+inference:
+  batch_size: 1
+  dim_t: 1101
+  num_overlap: 2

melbandroformers/karaoke/karaokegabox_1750911344.yaml

@@ -0,0 +1,72 @@
+audio:
+  chunk_size: 485100
+  dim_f: 1024
+  dim_t: 256
+  hop_length: 441
+  n_fft: 2048
+  num_channels: 2
+  sample_rate: 44100
+  min_mean_abs: 0.000
+
+model:
+  dim: 384
+  depth: 6
+  stereo: true
+  num_stems: 1
+  time_transformer_depth: 1
+  freq_transformer_depth: 1
+  num_bands: 60
+  dim_head: 64
+  heads: 8
+  attn_dropout: 0
+  ff_dropout: 0
+  flash_attn: true
+  dim_freqs_in: 1025
+  sample_rate: 44100  # needed for mel filter bank from librosa
+  stft_n_fft: 2048
+  stft_hop_length: 441
+  stft_win_length: 2048
+  stft_normalized: false
+  mask_estimator_depth: 2
+  multi_stft_resolution_loss_weight: 1.0
+  multi_stft_resolutions_window_sizes: !!python/tuple
+    - 4096
+    - 2048
+    - 1024
+    - 512
+    - 256
+  multi_stft_hop_size: 147
+  multi_stft_normalized: true
+
+training:
+  batch_size: 1
+  gradient_accumulation_steps: 1
+  grad_clip: 0
+  instruments:
+    - Vocals
+    - Instrumental
+  lr: 0.0005
+  patience: 2
+  reduce_factor: 0.95
+  target_instrument: Vocals
+  num_epochs: 1000
+  num_steps: 1000
+  augmentation: false # enable augmentations by audiomentations and pedalboard
+  augmentation_type:
+  use_mp3_compress: false # Deprecated
+  augmentation_mix: false # Mix several stems of the same type with some probability
+  augmentation_loudness: false # randomly change loudness of each stem
+  augmentation_loudness_type: 1 # Type 1 or 2
+  augmentation_loudness_min: 0
+  augmentation_loudness_max: 0
+  q: 0.95
+  coarse_loss_clip: false
+  ema_momentum: 0.999
+  optimizer: adamw
+  other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
+  use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
+
+inference:
+  batch_size: 1
+  dim_t: 1101
+  num_overlap: 8

melbandroformers/vocals/voc_Fv3.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49d81446b34a7848446efde7898b25bdc32fe872c2393617acb5356649f7ea93
+size 913026650

melbandroformers/vocals/voc_fv4.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a9657de5fd3ed87ad4fd1a9d2069743ecb33424836973ad0f3288e2a64e90bc
+size 913026650

melbandroformers/vocals/voc_fv5.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ede0504ddc55cb44b966a8212dac75a364f8157974cc40c8e92b9f5d4f17ce2
+size 913026650

melbandroformers/vocals/voc_gabox.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff802a67501fac70587c3ff4e8dbc89c2558e7d8911c92222dfea2aaac208517
+size 913026650

melbandroformers/vocals/voc_gabox.yaml

@@ -0,0 +1,51 @@
+audio:
+  chunk_size: 352800
+  dim_f: 1024
+  dim_t: 256
+  hop_length: 441
+  n_fft: 2048
+  num_channels: 2
+  sample_rate: 44100
+  min_mean_abs: 0.001
+
+model:
+  dim: 384
+  depth: 6
+  stereo: true
+  num_stems: 1
+  time_transformer_depth: 1
+  freq_transformer_depth: 1
+  num_bands: 60
+  dim_head: 64
+  heads: 8
+  attn_dropout: 0
+  ff_dropout: 0
+  flash_attn: True
+  dim_freqs_in: 1025
+  sample_rate: 44100  # needed for mel filter bank from librosa
+  stft_n_fft: 2048
+  stft_hop_length: 441
+  stft_win_length: 2048
+  stft_normalized: False
+  mask_estimator_depth: 2
+  multi_stft_resolution_loss_weight: 1.0
+  multi_stft_resolutions_window_sizes: !!python/tuple
+  - 4096
+  - 2048
+  - 1024
+  - 512
+  - 256
+  multi_stft_hop_size: 147
+  multi_stft_normalized: False
+
+training:
+  instruments:
+  - Vocals
+  - Instrumental
+  target_instrument: Vocals
+
+inference:
+  batch_size: 1
+  dim_t: 1101
+  num_overlap: 1
+  chunk_size: 352800

melbandroformers/vocals/voc_gaboxFv2.ckpt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2888813aa5b519941fa8548efc5a4331d63c61909007eb17fe95c367be230196
+size 913026650