diff --git a/.gitattributes b/.gitattributes index 66478df75bbf61cea445488a285a3b439ee5c1e7..f6339b0908ad95683fb80e6340e98c24ebd12c81 100644 --- a/.gitattributes +++ b/.gitattributes @@ -34,3 +34,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text KUIELab-MDX-Net.[[:space:]]A[[:space:]]Two-Stream[[:space:]]Neural[[:space:]]Network[[:space:]]for[[:space:]]Music[[:space:]]Demixing.pdf filter=lfs diff=lfs merge=lfs -text +models/audio_separation/Audio_Examples/01-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text +models/audio_separation/Audio_Examples/04-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text +models/audio_separation/Audio_Examples/05-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text diff --git a/models/audio-separation (melmass)/.gitattributes b/models/audio-separation (melmass)/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..dab9a4e17afd2ef39d90ccb0b40ef2786fe77422 --- /dev/null +++ b/models/audio-separation (melmass)/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/models/audio-separation (melmass)/README.md b/models/audio-separation (melmass)/README.md new file mode 100644 index 0000000000000000000000000000000000000000..7b95401dc46245ac339fc25059d4a56d90b4cde5 --- /dev/null +++ b/models/audio-separation (melmass)/README.md @@ -0,0 +1,3 @@ +--- +license: apache-2.0 +--- diff --git a/models/audio-separation (melmass)/model_vocals_mdx23c_sdr_10.17.safetensors b/models/audio-separation (melmass)/model_vocals_mdx23c_sdr_10.17.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8427a77d50c64f3f26a8f15afa4e4221c0528b17 --- /dev/null +++ b/models/audio-separation (melmass)/model_vocals_mdx23c_sdr_10.17.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6f2a46e602f58b3c849ac6b2596f08636716ab26308b4f2e15c9c986b2cf2ebe +size 447998288 diff --git a/models/audio-separation (melmass)/source.txt b/models/audio-separation (melmass)/source.txt new file mode 100644 index 0000000000000000000000000000000000000000..3c15c7ecca28129a18918ff09f1c99899730331c --- /dev/null +++ b/models/audio-separation (melmass)/source.txt @@ -0,0 +1 @@ +https://huggingface.co/melmass/audio-separation \ No newline at end of file diff --git a/models/audio-separation-models/.gitattributes b/models/audio-separation-models/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..dab9a4e17afd2ef39d90ccb0b40ef2786fe77422 --- /dev/null +++ b/models/audio-separation-models/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/models/audio-separation-models/mdx23c-dereverb/config_dereverb_mdx23c.yaml b/models/audio-separation-models/mdx23c-dereverb/config_dereverb_mdx23c.yaml new file mode 100644 index 0000000000000000000000000000000000000000..a35c93b0b5e83ea36d42d60af0bb8bee40541c51 --- /dev/null +++ b/models/audio-separation-models/mdx23c-dereverb/config_dereverb_mdx23c.yaml @@ -0,0 +1,135 @@ +audio: + chunk_size: 261120 + dim_f: 4096 + dim_t: 256 + hop_length: 1024 + n_fft: 8192 + num_channels: 2 + sample_rate: 44100 + min_mean_abs: 0.001 + +model: + act: gelu + bottleneck_factor: 4 + growth: 128 + norm: InstanceNorm + num_blocks_per_scale: 2 + num_channels: 128 + num_scales: 5 + num_subbands: 4 + scale: + - 2 + - 2 + +training: + batch_size: 2 + gradient_accumulation_steps: 1 + grad_clip: 0 + instruments: + - dry + - other + lr: 1.0e-06 + patience: 4 + reduce_factor: 0.93 + target_instrument: null + num_epochs: 40 + num_steps: 1000 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + optimizer: adamw + read_metadata_procs: 8 # Number of processes to use during metadata reading for dataset. Can speed up metadata generation + 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 + +augmentations: + enable: false # enable or disable all augmentations (to fast disable if needed) + loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max) + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3) + mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02) + - 0.2 + - 0.02 + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + + all: + channel_shuffle: 0.5 # Set 0 or lower to disable + random_inverse: 0.05 # inverse track (better lower probability) + random_polarity: 0.5 # polarity change (multiply waveform to -1) + + # pedalboard chorus block + pedalboard_chorus: 0.001 + pedalboard_chorus_rate_hz_min: 1.0 + pedalboard_chorus_rate_hz_max: 7.0 + pedalboard_chorus_depth_min: 0.25 + pedalboard_chorus_depth_max: 0.95 + pedalboard_chorus_centre_delay_ms_min: 3 + pedalboard_chorus_centre_delay_ms_max: 10 + pedalboard_chorus_feedback_min: 0.0 + pedalboard_chorus_feedback_max: 0.01 + pedalboard_chorus_mix_min: 0.1 + pedalboard_chorus_mix_max: 0.9 + + # pedalboard phazer block + pedalboard_phazer: 0.001 + pedalboard_phazer_rate_hz_min: 1.0 + pedalboard_phazer_rate_hz_max: 10.0 + pedalboard_phazer_depth_min: 0.25 + pedalboard_phazer_depth_max: 0.95 + pedalboard_phazer_centre_frequency_hz_min: 200 + pedalboard_phazer_centre_frequency_hz_max: 12000 + pedalboard_phazer_feedback_min: 0.0 + pedalboard_phazer_feedback_max: 0.5 + pedalboard_phazer_mix_min: 0.1 + pedalboard_phazer_mix_max: 0.9 + + # pedalboard pitch shift block + pedalboard_pitch_shift: 0.01 + pedalboard_pitch_shift_semitones_min: -7 + pedalboard_pitch_shift_semitones_max: 7 + + # pedalboard resample block + pedalboard_resample: 0.001 + pedalboard_resample_target_sample_rate_min: 4000 + pedalboard_resample_target_sample_rate_max: 44100 + + mp3_compression_min_bitrate: 32 + mp3_compression_max_bitrate: 320 + mp3_compression_backend: "lameenc" + + dry: + # pedalboard distortion block + pedalboard_distortion: 0.001 + pedalboard_distortion_drive_db_min: 1.0 + pedalboard_distortion_drive_db_max: 25.0 + + tanh_distortion: 0.05 + tanh_distortion_min: 0.1 + tanh_distortion_max: 0.7 + # pedalboard bitcrash block + pedalboard_bitcrash: 0.005 + pedalboard_bitcrash_bit_depth_min: 4 + pedalboard_bitcrash_bit_depth_max: 16 + + seven_band_parametric_eq: 0.24 + seven_band_parametric_eq_min_gain_db: -9 + seven_band_parametric_eq_max_gain_db: 9 + + gaussian_noise: 0.005 + gaussian_noise_min_amplitude: 0.001 + gaussian_noise_max_amplitude: 0.01 + + time_stretch: 0.01 + time_stretch_min_rate: 0.8 + time_stretch_max_rate: 1.25 + other: + seven_band_parametric_eq: 0.24 + seven_band_parametric_eq_min_gain_db: -9 + seven_band_parametric_eq_max_gain_db: 9 + +inference: + batch_size: 2 + dim_t: 256 + num_overlap: 4 \ No newline at end of file diff --git a/models/audio-separation-models/mdx23c-dereverb/dereverb_mdx23c_sdr_6.9096.ckpt b/models/audio-separation-models/mdx23c-dereverb/dereverb_mdx23c_sdr_6.9096.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..dca4c63709a57308193ba52a2d3309efed5915bc --- /dev/null +++ b/models/audio-separation-models/mdx23c-dereverb/dereverb_mdx23c_sdr_6.9096.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eae2471b707758d74db38ac1b1d5800e12f57c4e9d1ebbb2faf004b8e086e914 +size 448098867 diff --git a/models/audio-separation-models/source.txt b/models/audio-separation-models/source.txt new file mode 100644 index 0000000000000000000000000000000000000000..64967bbc3fe3e424b8c55fba9e138f065290cfe5 --- /dev/null +++ b/models/audio-separation-models/source.txt @@ -0,0 +1 @@ +https://huggingface.co/shiromiya/audio-separation-models \ No newline at end of file diff --git a/models/audio-separator-models/.gitattributes b/models/audio-separator-models/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..dab9a4e17afd2ef39d90ccb0b40ef2786fe77422 --- /dev/null +++ b/models/audio-separator-models/.gitattributes @@ -0,0 +1,35 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text diff --git a/models/audio-separator-models/README.md b/models/audio-separator-models/README.md new file mode 100644 index 0000000000000000000000000000000000000000..7be5fc7f47d5db027d120b8024982df93db95b74 --- /dev/null +++ b/models/audio-separator-models/README.md @@ -0,0 +1,3 @@ +--- +license: mit +--- diff --git a/models/audio-separator-models/mdx-net/Kim_Inst.onnx b/models/audio-separator-models/mdx-net/Kim_Inst.onnx new file mode 100644 index 0000000000000000000000000000000000000000..a1cfbc17713f1caef9d11696002bb78aae781ff2 --- /dev/null +++ b/models/audio-separator-models/mdx-net/Kim_Inst.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:86b1940e7122fbdd2beadc65507cbff6c352d79012a8a7e60d56db98532af5f7 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/Kim_Vocal_1.onnx b/models/audio-separator-models/mdx-net/Kim_Vocal_1.onnx new file mode 100644 index 0000000000000000000000000000000000000000..4d6e31ac4c3e53021d67bf71f7aa576bbff391b5 --- /dev/null +++ b/models/audio-separator-models/mdx-net/Kim_Vocal_1.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f313140ef8fecc3041881b60ecb993d985a0281a138b2fb634aa8901aebc38cb +size 66759214 diff --git a/models/audio-separator-models/mdx-net/Kim_Vocal_2.onnx b/models/audio-separator-models/mdx-net/Kim_Vocal_2.onnx new file mode 100644 index 0000000000000000000000000000000000000000..89d344b3926e4d0d7ee41b78e7f387558ec2acdc --- /dev/null +++ b/models/audio-separator-models/mdx-net/Kim_Vocal_2.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce74ef3b6a6024ce44211a07be9cf8bc6d87728cc852a68ab34eb8e58cde9c8b +size 66759214 diff --git a/models/audio-separator-models/mdx-net/Reverb_HQ_By_FoxJoy.onnx b/models/audio-separator-models/mdx-net/Reverb_HQ_By_FoxJoy.onnx new file mode 100644 index 0000000000000000000000000000000000000000..0af4086572259731e9237a7d5acc2254f2d30cf5 --- /dev/null +++ b/models/audio-separator-models/mdx-net/Reverb_HQ_By_FoxJoy.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:233bb5c6aaa365e568659a0a81211746fa881f8f47f82d9e864fce1f7692db80 +size 66780123 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_1.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_1.onnx new file mode 100644 index 0000000000000000000000000000000000000000..86732ec554ea5624474d9d109f6a72b3570fdb34 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_1.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6ca53f94b7a0cbb04fcfcc8f3ea5ec1ae22cd8ad044f5e673588859f83976f5e +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_2.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_2.onnx new file mode 100644 index 0000000000000000000000000000000000000000..3d7d5ea74c2caf49634f847217378bb626e2e77a --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_2.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e3a96a664d28b52db9def0a9cae9a16dbb524d8325bfe8f0ac64ac5d231456bc +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_3.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_3.onnx new file mode 100644 index 0000000000000000000000000000000000000000..548cacd28b73e2f76dde5f4281f1e0d0dcf537e5 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_3.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b7834e2972158d8c9864e7376e3a7d084079c80a23f38dc31c4b0a4e901a1cb +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_1.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_1.onnx new file mode 100644 index 0000000000000000000000000000000000000000..3302ace57202067fd1e9c709d76bc8d347658f5d --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_1.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:38a045c4ded87e3bf97b609ec5be7910e8a7cecec455f507227ab12b5e29f7f9 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_2.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_2.onnx new file mode 100644 index 0000000000000000000000000000000000000000..1340d10933420baf427d06cbf9a51ac7903b5398 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_2.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:197f8ab296df850f961e68c595f6649acb7d9e621b5600b460f3458967299112 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_3.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_3.onnx new file mode 100644 index 0000000000000000000000000000000000000000..f2b6b241d9246f392067e0717c8b252857022e6c --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_3.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:317554b07fe1ea5279a77f2b1520a41ea4b93432560c4ffd08792c30fddf9adc +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_4.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_4.onnx new file mode 100644 index 0000000000000000000000000000000000000000..45b11285dea638eeecfe4487dfc39775da075b14 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_4.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c4b5b9b05090fdf238f38ba5046813982d50e2a652e9cb3324ea79720c3c9c8 +size 59074342 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_5.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_5.onnx new file mode 100644 index 0000000000000000000000000000000000000000..e674c464504ee1cddb1371129a4314eacabbca64 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_HQ_5.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:811cb24095d865763752310848b7ec86aeede0626cb05749ab35350e46897000 +size 59074342 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_Main.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_Main.onnx new file mode 100644 index 0000000000000000000000000000000000000000..3c1ab77a4f5874c013ff0269cc926fcc9a7ab466 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_Main.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ab401dfe4a548b87deb64f975294bd56ff946aa32903f53b4b24bb13b2cce1e +size 52786726 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_full_292.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_full_292.onnx new file mode 100644 index 0000000000000000000000000000000000000000..9f6c60def7873f03a26391afabb238103fc5c318 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Inst_full_292.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:020f6b65fa219fb7c285e4f3fc2863bf22daf03c4c93e547b6d13d5f2757a7ec +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET-Voc_FT.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Voc_FT.onnx new file mode 100644 index 0000000000000000000000000000000000000000..3048949a6f427d212f310e9a13494306da6e00ec --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET-Voc_FT.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:534b2070fcc7df514b13ef660dc8cbb328679c2374d04354a5c42bb14ecce111 +size 66762490 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Crowd_HQ_1.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Crowd_HQ_1.onnx new file mode 100644 index 0000000000000000000000000000000000000000..466c3fa69b05f5b27c19cc11eb23c99909d2a4d0 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Crowd_HQ_1.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:313b7bf869c411fdafe005cf0d5a635c405cb3d0df137178a64091952d75225c +size 59074342 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_187_beta.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_187_beta.onnx new file mode 100644 index 0000000000000000000000000000000000000000..b2ff188d7a8ea6ed25bcf1916359853b8fd0cb8f --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_187_beta.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c74566f3c3033cacba996328b2ee90bf77ef79ea6c35b7841df183b7906f54a5 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_82_beta.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_82_beta.onnx new file mode 100644 index 0000000000000000000000000000000000000000..f91e15a00d9d805623d70dc7d95b69e2e8329ba2 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_82_beta.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c6c268302f09ab53687072618e056a611272a7e2c3fd9b3b59164da152f3588e +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_90_beta.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_90_beta.onnx new file mode 100644 index 0000000000000000000000000000000000000000..db0d80d538ff6ecf3a15cabaa0c84500ff0b5ccc --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Inst_90_beta.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d902868a46575aea6ee2335736ff3b53faf497a6bdaa1b864e0fd84eb1b42a5 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_340.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_340.onnx new file mode 100644 index 0000000000000000000000000000000000000000..75ef0024acd46a2900ea78e948325d044ec7580a --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_340.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:78792633b4007755af12ecde20f709b4f0b99563b1d25fe0a501ed2122aff218 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_390.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_390.onnx new file mode 100644 index 0000000000000000000000000000000000000000..a752391e7cfb5b134ebef388734cd1da9eb5dfb5 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_390.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:286c4f0847ca837e2c3f4c4058f756d5f150cbf080506aa6f33a2847aba92e8c +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_406.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_406.onnx new file mode 100644 index 0000000000000000000000000000000000000000..c0f58423f248611332f6a3c33382d0d426c67449 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_406.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f158816a44eef1f0ba0f48b813cbfcf460ed1c70a754af3609ade44aaf7d1b23 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_427.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_427.onnx new file mode 100644 index 0000000000000000000000000000000000000000..f6731f9dcc99c5b1ad0af85dd845280e833acf46 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_427.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:95275802a27801b97e3c0552b6eaa69f9bb3bd7df53cdf0536cce0a753f702cc +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_438.onnx b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_438.onnx new file mode 100644 index 0000000000000000000000000000000000000000..a6a17015453da90871426317c547193196bd6f1e --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR-MDX-NET_Main_438.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d5e1ad93587a163a0987a0168b99a2ad875c0d9bfc3afb596b7c36b09c7f5c26 +size 66759214 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_1_9703.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_1_9703.onnx new file mode 100644 index 0000000000000000000000000000000000000000..50d00b7e34e7763954283b9fc13f2d903072be03 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_1_9703.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:229ad3bb96a037e89d8ed86732d6d3675856e6a07c3e3f02896eac01ec7ee4be +size 29704436 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_2_9682.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_2_9682.onnx new file mode 100644 index 0000000000000000000000000000000000000000..1d4c430b6d6a8a7d8abea9bccac1959e4ea69ab0 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_2_9682.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1deb7295acd3206bc9582a5d92f1b0a74bf3f41c7c1fb78a0ac0123cde4372db +size 29704436 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_3_9662.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_3_9662.onnx new file mode 100644 index 0000000000000000000000000000000000000000..140b8e3eb273df75384c691462998774b3928a52 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_3_9662.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e02220e80d8253f4c2209f8924298b2b686bbdf2868b788ff5500fb9bd94aadc +size 29704436 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_9482.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_9482.onnx new file mode 100644 index 0000000000000000000000000000000000000000..c40c9257378536385f625a75f7bc74cb35c1eaa5 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_9482.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f4f365207c56deb115bceedff3ad8fe98a751c745f9e370cecec6226b8b47184 +size 29704436 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA.onnx new file mode 100644 index 0000000000000000000000000000000000000000..4bfb51872cc19150179df558fb7212e5359427ea --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e3167c87333a48548413e972a286bf40bf5694001d2853861eb1435953f02d63 +size 29704436 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA_2.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA_2.onnx new file mode 100644 index 0000000000000000000000000000000000000000..18aeb6a8d05b84fa74fdd0de7ed917f6df273ebd --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_KARA_2.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bf32e15105a09c0f7dddd2b67346146334d6f3ecb399ed7638eba2ab07cbf5f4 +size 52786726 diff --git a/models/audio-separator-models/mdx-net/UVR_MDXNET_Main.onnx b/models/audio-separator-models/mdx-net/UVR_MDXNET_Main.onnx new file mode 100644 index 0000000000000000000000000000000000000000..d9645fd239fd211619aabeb431d3093a45ed8b85 --- /dev/null +++ b/models/audio-separator-models/mdx-net/UVR_MDXNET_Main.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8289784cda38543ff431add4070662813311a8cccfc0112ca82f76d9dba2b4ca +size 66759214 diff --git a/models/audio-separator-models/mdx-net/kuielab_a_bass.onnx b/models/audio-separator-models/mdx-net/kuielab_a_bass.onnx new file mode 100644 index 0000000000000000000000000000000000000000..0fe816090f69e7198dc780025d2c37b9dc9a272f --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_a_bass.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0c3e77b9963185b1ea6bb46a4b8924137d9370fc1ccdefec7b1b416ef550dcaa +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_a_drums.onnx b/models/audio-separator-models/mdx-net/kuielab_a_drums.onnx new file mode 100644 index 0000000000000000000000000000000000000000..e29380d6f0c92fe7a38a201709342a56931489df --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_a_drums.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:40f586b7091934dd6f5563f0cba8f14bad57ce88440da1098bf388ea716c2901 +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_a_other.onnx b/models/audio-separator-models/mdx-net/kuielab_a_other.onnx new file mode 100644 index 0000000000000000000000000000000000000000..7f2c3285075afc5e9ed9ba2fd6882d8df9921763 --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_a_other.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7b67a1dcb5f232153528c59960b4c7bf8dc736b8114de360af0e719633f53358 +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_a_vocals.onnx b/models/audio-separator-models/mdx-net/kuielab_a_vocals.onnx new file mode 100644 index 0000000000000000000000000000000000000000..8d80e6d5ea023a73a002c15005551a37c7cfa21c --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_a_vocals.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:daba83c2ee1afee9139766ad64c9b6808d6b6f092fff04bed3338be50baac721 +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_b_bass.onnx b/models/audio-separator-models/mdx-net/kuielab_b_bass.onnx new file mode 100644 index 0000000000000000000000000000000000000000..1670d9993ab4080cc702ba397b629caae20ac699 --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_b_bass.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b4b7080fe501d0bece62076c5d4eda4d6590c5207ed78ec84a57bac0740a061d +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_b_drums.onnx b/models/audio-separator-models/mdx-net/kuielab_b_drums.onnx new file mode 100644 index 0000000000000000000000000000000000000000..a00f9c84fd964d45c2cdd0470ff9e4b5a7f4ad27 --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_b_drums.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a6fecee758059b33ed99f6dabba297439b3e7cacfac4b1097bd324aff8052208 +size 21930313 diff --git a/models/audio-separator-models/mdx-net/kuielab_b_other.onnx b/models/audio-separator-models/mdx-net/kuielab_b_other.onnx new file mode 100644 index 0000000000000000000000000000000000000000..e17baa7f40b3dbd1299ee7ba7e6db461682e8320 --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_b_other.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b0d0b63950ac332333fea2d58f68c92fd3ab0aae071398c2a8beeae1ad15b655 +size 29703204 diff --git a/models/audio-separator-models/mdx-net/kuielab_b_vocals.onnx b/models/audio-separator-models/mdx-net/kuielab_b_vocals.onnx new file mode 100644 index 0000000000000000000000000000000000000000..683e555b8a91e3035cc24bee5b47b7347390eb8a --- /dev/null +++ b/models/audio-separator-models/mdx-net/kuielab_b_vocals.onnx @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9b7dcb9d878acb0f3e64ff3fd27750faae96577013f6d50f5996875bf4250713 +size 29703204 diff --git a/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ.ckpt b/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..d19b34ba845c464fd607a4ce4c9dd3f794d0bd23 --- /dev/null +++ b/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:49d51472769e34a2501cd1da782346a3212555c3a5619fc2c53507445528d816 +size 448101203 diff --git a/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ_2.ckpt b/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ_2.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..486c55bd1538c5471f02f9e08af5ceb8baec08e2 --- /dev/null +++ b/models/audio-separator-models/mdx23c/MDX23C-8KFFT-InstVoc_HQ_2.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7d960d8e40a458120412c1bd807e013d2dbca7b959cc9da2bbcb0eb203d1daea +size 448093770 diff --git a/models/audio-separator-models/mdx23c/MDX23C-De-Reverb-aufr33-jarredou.ckpt b/models/audio-separator-models/mdx23c/MDX23C-De-Reverb-aufr33-jarredou.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..dca4c63709a57308193ba52a2d3309efed5915bc --- /dev/null +++ b/models/audio-separator-models/mdx23c/MDX23C-De-Reverb-aufr33-jarredou.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eae2471b707758d74db38ac1b1d5800e12f57c4e9d1ebbb2faf004b8e086e914 +size 448098867 diff --git a/models/audio-separator-models/mdx23c/MDX23C-DrumSep-aufr33-jarredou.ckpt b/models/audio-separator-models/mdx23c/MDX23C-DrumSep-aufr33-jarredou.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..2f97198d19fcae11a05adbd6b2a746e1f466833f --- /dev/null +++ b/models/audio-separator-models/mdx23c/MDX23C-DrumSep-aufr33-jarredou.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d2a4aa53eb584d21eead358a4e66d1882ad182911be018f052b5da73be9096d0 +size 437652699 diff --git a/models/audio-separator-models/mdx23c/MDX23C_D1581.ckpt b/models/audio-separator-models/mdx23c/MDX23C_D1581.ckpt new file mode 100644 index 0000000000000000000000000000000000000000..62aeea1ec485760c60a560dbcc87e44995964d23 --- /dev/null +++ b/models/audio-separator-models/mdx23c/MDX23C_D1581.ckpt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7d4d2137f12764950205b095da20032fef1d41f077bacc8582f20ed40e8cb28 +size 183379219 diff --git a/models/audio-separator-models/mdx23c/config_dereverb_mdx23c.yaml b/models/audio-separator-models/mdx23c/config_dereverb_mdx23c.yaml new file mode 100644 index 0000000000000000000000000000000000000000..32ab7d8cc97b51e94096c6cdaabaee3e987e8830 --- /dev/null +++ b/models/audio-separator-models/mdx23c/config_dereverb_mdx23c.yaml @@ -0,0 +1,135 @@ +audio: + chunk_size: 261120 + dim_f: 4096 + dim_t: 256 + hop_length: 1024 + n_fft: 8192 + num_channels: 2 + sample_rate: 44100 + min_mean_abs: 0.001 + +model: + act: gelu + bottleneck_factor: 4 + growth: 128 + norm: InstanceNorm + num_blocks_per_scale: 2 + num_channels: 128 + num_scales: 5 + num_subbands: 4 + scale: + - 2 + - 2 + +training: + batch_size: 2 + gradient_accumulation_steps: 1 + grad_clip: 0 + instruments: + - dry + - No dry + lr: 1.0e-06 + patience: 4 + reduce_factor: 0.93 + target_instrument: null + num_epochs: 40 + num_steps: 1000 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + optimizer: adamw + read_metadata_procs: 8 # Number of processes to use during metadata reading for dataset. Can speed up metadata generation + 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 + +augmentations: + enable: false # enable or disable all augmentations (to fast disable if needed) + loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max) + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3) + mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02) + - 0.2 + - 0.02 + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + + all: + channel_shuffle: 0.5 # Set 0 or lower to disable + random_inverse: 0.05 # inverse track (better lower probability) + random_polarity: 0.5 # polarity change (multiply waveform to -1) + + # pedalboard chorus block + pedalboard_chorus: 0.001 + pedalboard_chorus_rate_hz_min: 1.0 + pedalboard_chorus_rate_hz_max: 7.0 + pedalboard_chorus_depth_min: 0.25 + pedalboard_chorus_depth_max: 0.95 + pedalboard_chorus_centre_delay_ms_min: 3 + pedalboard_chorus_centre_delay_ms_max: 10 + pedalboard_chorus_feedback_min: 0.0 + pedalboard_chorus_feedback_max: 0.01 + pedalboard_chorus_mix_min: 0.1 + pedalboard_chorus_mix_max: 0.9 + + # pedalboard phazer block + pedalboard_phazer: 0.001 + pedalboard_phazer_rate_hz_min: 1.0 + pedalboard_phazer_rate_hz_max: 10.0 + pedalboard_phazer_depth_min: 0.25 + pedalboard_phazer_depth_max: 0.95 + pedalboard_phazer_centre_frequency_hz_min: 200 + pedalboard_phazer_centre_frequency_hz_max: 12000 + pedalboard_phazer_feedback_min: 0.0 + pedalboard_phazer_feedback_max: 0.5 + pedalboard_phazer_mix_min: 0.1 + pedalboard_phazer_mix_max: 0.9 + + # pedalboard pitch shift block + pedalboard_pitch_shift: 0.01 + pedalboard_pitch_shift_semitones_min: -7 + pedalboard_pitch_shift_semitones_max: 7 + + # pedalboard resample block + pedalboard_resample: 0.001 + pedalboard_resample_target_sample_rate_min: 4000 + pedalboard_resample_target_sample_rate_max: 44100 + + mp3_compression_min_bitrate: 32 + mp3_compression_max_bitrate: 320 + mp3_compression_backend: "lameenc" + + dry: + # pedalboard distortion block + pedalboard_distortion: 0.001 + pedalboard_distortion_drive_db_min: 1.0 + pedalboard_distortion_drive_db_max: 25.0 + + tanh_distortion: 0.05 + tanh_distortion_min: 0.1 + tanh_distortion_max: 0.7 + # pedalboard bitcrash block + pedalboard_bitcrash: 0.005 + pedalboard_bitcrash_bit_depth_min: 4 + pedalboard_bitcrash_bit_depth_max: 16 + + seven_band_parametric_eq: 0.24 + seven_band_parametric_eq_min_gain_db: -9 + seven_band_parametric_eq_max_gain_db: 9 + + gaussian_noise: 0.005 + gaussian_noise_min_amplitude: 0.001 + gaussian_noise_max_amplitude: 0.01 + + time_stretch: 0.01 + time_stretch_min_rate: 0.8 + time_stretch_max_rate: 1.25 + other: + seven_band_parametric_eq: 0.24 + seven_band_parametric_eq_min_gain_db: -9 + seven_band_parametric_eq_max_gain_db: 9 + +inference: + batch_size: 2 + dim_t: 256 + num_overlap: 4 \ No newline at end of file diff --git a/models/audio-separator-models/mdx23c/config_drumsep_mdx23c.yaml b/models/audio-separator-models/mdx23c/config_drumsep_mdx23c.yaml new file mode 100644 index 0000000000000000000000000000000000000000..7b06ddb77b6d2df982922cabaa9805a189f4c6c6 --- /dev/null +++ b/models/audio-separator-models/mdx23c/config_drumsep_mdx23c.yaml @@ -0,0 +1,87 @@ +audio: + chunk_size: 130560 + dim_f: 1024 + dim_t: 256 + hop_length: 512 + n_fft: 2048 + num_channels: 2 + sample_rate: 44100 + min_mean_abs: 0.001 + +model: + act: gelu + bottleneck_factor: 4 + growth: 128 + norm: InstanceNorm + num_blocks_per_scale: 2 + num_channels: 128 + num_scales: 5 + num_subbands: 4 + scale: + - 2 + - 2 + +training: + batch_size: 12 + gradient_accumulation_steps: 1 + grad_clip: 0 + instruments: + - kick + - snare + - toms + - hh + - ride + - crash + lr: 9.0e-05 + patience: 30 + reduce_factor: 0.95 + target_instrument: null + num_epochs: 1000 + num_steps: 1268 + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 + optimizer: adam + 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 + +augmentations: + enable: true # enable or disable all augmentations (to fast disable if needed) + loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max) + loudness_min: 0.5 + loudness_max: 1.5 + mixup: true # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3) + mixup_probs: !!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02) + - 0.2 + - 0.02 + mixup_loudness_min: 0.5 + mixup_loudness_max: 1.5 + + # apply mp3 compression to mixture only (emulate downloading mp3 from internet) + mp3_compression_on_mixture: 0.0 + mp3_compression_on_mixture_bitrate_min: 32 + mp3_compression_on_mixture_bitrate_max: 320 + mp3_compression_on_mixture_backend: "lameenc" + + all: + channel_shuffle: 0.5 # Set 0 or lower to disable + random_inverse: 0.01 # inverse track (better lower probability) + random_polarity: 0.5 # polarity change (multiply waveform to -1) + mp3_compression: 0.0 + mp3_compression_min_bitrate: 32 + mp3_compression_max_bitrate: 320 + mp3_compression_backend: "lameenc" + pitch_shift: 0.1 + pitch_shift_min_semitones: -3 + pitch_shift_max_semitones: 3 + seven_band_parametric_eq: 0.5 + seven_band_parametric_eq_min_gain_db: -6 + seven_band_parametric_eq_max_gain_db: 6 + tanh_distortion: 0.2 + tanh_distortion_min: 0.1 + tanh_distortion_max: 0.5 + +inference: + batch_size: 1 + dim_t: 256 + num_overlap: 4 \ No newline at end of file diff --git a/models/audio-separator-models/mdx23c/model_2_stem_061321.yaml b/models/audio-separator-models/mdx23c/model_2_stem_061321.yaml new file mode 100644 index 0000000000000000000000000000000000000000..332c497db4d65fe2e2a6523e7b6a63c93f7fc30e --- /dev/null +++ b/models/audio-separator-models/mdx23c/model_2_stem_061321.yaml @@ -0,0 +1,36 @@ +audio: + chunk_size: 260096 + dim_f: 4096 + dim_t: 256 + hop_length: 2048 + n_fft: 12288 + num_channels: 2 + sample_rate: 44100 + min_mean_abs: 0.001 +model: + act: gelu + bottleneck_factor: 4 + growth: 64 + norm: InstanceNorm + num_blocks_per_scale: 2 + num_channels: 128 + num_scales: 5 + num_subbands: 4 + scale: + - 2 + - 2 + name: epoch_10.ckpt +training: + batch_size: 16 + grad_clip: 0 + instruments: + - Vocals + - Instrumental + lr: 5.0e-05 + target_instrument: null + num_epochs: 100 + num_steps: 1000 +inference: + batch_size: 1 + dim_t: 256 + num_overlap: 8 diff --git a/models/audio-separator-models/mdx23c/model_2_stem_full_band_8k.yaml b/models/audio-separator-models/mdx23c/model_2_stem_full_band_8k.yaml new file mode 100644 index 0000000000000000000000000000000000000000..66862b034aeab482c925aa2359be3dcbe276527a --- /dev/null +++ b/models/audio-separator-models/mdx23c/model_2_stem_full_band_8k.yaml @@ -0,0 +1,43 @@ +audio: + chunk_size: 261120 + dim_f: 4096 + dim_t: 256 + hop_length: 1024 + n_fft: 8192 + num_channels: 2 + sample_rate: 44100 + min_mean_abs: 0.001 +model: + act: gelu + bottleneck_factor: 4 + growth: 128 + norm: InstanceNorm + num_blocks_per_scale: 2 + num_channels: 128 + num_scales: 5 + num_subbands: 4 + scale: + - 2 + - 2 +training: + batch_size: 6 + grad_clip: 0 + instruments: + - Vocals + - Instrumental + lr: 1.0e-05 + patience: 2 + reduce_factor: 0.95 + target_instrument: null + num_epochs: 1000 + num_steps: 1000 + augmentation: 1 + augmentation_type: simple1 + augmentation_mix: true + q: 0.95 + coarse_loss_clip: true + ema_momentum: 0.999 +inference: + batch_size: 1 + dim_t: 256 + num_overlap: 8 \ No newline at end of file diff --git a/models/audio_separation/.gitattributes b/models/audio_separation/.gitattributes new file mode 100644 index 0000000000000000000000000000000000000000..8733933e846d6d91a8a0aaff65ed783af3abdb35 --- /dev/null +++ b/models/audio_separation/.gitattributes @@ -0,0 +1,38 @@ +*.7z filter=lfs diff=lfs merge=lfs -text +*.arrow filter=lfs diff=lfs merge=lfs -text +*.bin filter=lfs diff=lfs merge=lfs -text +*.bz2 filter=lfs diff=lfs merge=lfs -text +*.ckpt filter=lfs diff=lfs merge=lfs -text +*.ftz filter=lfs diff=lfs merge=lfs -text +*.gz filter=lfs diff=lfs merge=lfs -text +*.h5 filter=lfs diff=lfs merge=lfs -text +*.joblib filter=lfs diff=lfs merge=lfs -text +*.lfs.* filter=lfs diff=lfs merge=lfs -text +*.mlmodel filter=lfs diff=lfs merge=lfs -text +*.model filter=lfs diff=lfs merge=lfs -text +*.msgpack filter=lfs diff=lfs merge=lfs -text +*.npy filter=lfs diff=lfs merge=lfs -text +*.npz filter=lfs diff=lfs merge=lfs -text +*.onnx filter=lfs diff=lfs merge=lfs -text +*.ot filter=lfs diff=lfs merge=lfs -text +*.parquet filter=lfs diff=lfs merge=lfs -text +*.pb filter=lfs diff=lfs merge=lfs -text +*.pickle filter=lfs diff=lfs merge=lfs -text +*.pkl filter=lfs diff=lfs merge=lfs -text +*.pt filter=lfs diff=lfs merge=lfs -text +*.pth filter=lfs diff=lfs merge=lfs -text +*.rar filter=lfs diff=lfs merge=lfs -text +*.safetensors filter=lfs diff=lfs merge=lfs -text +saved_model/**/* filter=lfs diff=lfs merge=lfs -text +*.tar.* filter=lfs diff=lfs merge=lfs -text +*.tar filter=lfs diff=lfs merge=lfs -text +*.tflite filter=lfs diff=lfs merge=lfs -text +*.tgz filter=lfs diff=lfs merge=lfs -text +*.wasm filter=lfs diff=lfs merge=lfs -text +*.xz filter=lfs diff=lfs merge=lfs -text +*.zip filter=lfs diff=lfs merge=lfs -text +*.zst filter=lfs diff=lfs merge=lfs -text +*tfevents* filter=lfs diff=lfs merge=lfs -text +Audio_Examples/01-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text +Audio_Examples/04-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text +Audio_Examples/05-The_Symphony_of_Automation.mp3 filter=lfs diff=lfs merge=lfs -text diff --git a/models/audio_separation/Audio_Examples/01-The_Symphony_of_Automation.mp3 b/models/audio_separation/Audio_Examples/01-The_Symphony_of_Automation.mp3 new file mode 100644 index 0000000000000000000000000000000000000000..d765bea2639943e50ee27bd581700462628cc7d9 --- /dev/null +++ b/models/audio_separation/Audio_Examples/01-The_Symphony_of_Automation.mp3 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97dabae247ff7f40f6951e5ce7a4c3a1092063bafeeead3abf1abf0a9b7f63a4 +size 4057389 diff --git a/models/audio_separation/Audio_Examples/04-The_Symphony_of_Automation.mp3 b/models/audio_separation/Audio_Examples/04-The_Symphony_of_Automation.mp3 new file mode 100644 index 0000000000000000000000000000000000000000..1de42f33c6a67b867d786eb7b246a77a8f72f9bc --- /dev/null +++ b/models/audio_separation/Audio_Examples/04-The_Symphony_of_Automation.mp3 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4e184949f498addb0a666529bc0f79553ec79a5524459efb7360fe27bfd96376 +size 3229101 diff --git a/models/audio_separation/Audio_Examples/05-The_Symphony_of_Automation.mp3 b/models/audio_separation/Audio_Examples/05-The_Symphony_of_Automation.mp3 new file mode 100644 index 0000000000000000000000000000000000000000..0c62c59a822c47f0519da6b74f499a7171305452 --- /dev/null +++ b/models/audio_separation/Audio_Examples/05-The_Symphony_of_Automation.mp3 @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2e738279500ad726504fccb1a383b1616a1cc967e72951654a28e464af776be9 +size 4773357 diff --git a/models/audio_separation/Demucs/UVR_Demucs_Model_1.yaml b/models/audio_separation/Demucs/UVR_Demucs_Model_1.yaml new file mode 100644 index 0000000000000000000000000000000000000000..18faa91d34b98edd25819aa6ffce033dd4b96f15 --- /dev/null +++ b/models/audio_separation/Demucs/UVR_Demucs_Model_1.yaml @@ -0,0 +1,2 @@ +models: ['ebf34a2db'] +segment: 44 \ No newline at end of file diff --git a/models/audio_separation/Demucs/UVR_Demucs_Model_2.yaml b/models/audio_separation/Demucs/UVR_Demucs_Model_2.yaml new file mode 100644 index 0000000000000000000000000000000000000000..b5842cc71c6d9d1b7a77f56fe5efd74e66824ea3 --- /dev/null +++ b/models/audio_separation/Demucs/UVR_Demucs_Model_2.yaml @@ -0,0 +1,2 @@ +models: ['ebf34a2d'] +segment: 44 \ No newline at end of file diff --git a/models/audio_separation/Demucs/UVR_Demucs_Model_Bag.safetensors b/models/audio_separation/Demucs/UVR_Demucs_Model_Bag.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b69bfcdc668cba58940842f8022de0955af73ada --- /dev/null +++ b/models/audio_separation/Demucs/UVR_Demucs_Model_Bag.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5065ceb6cd1b4a637f16c26c1cfa2aee1202a5f90a08829c700284f2c44a9fb0 +size 334624768 diff --git a/models/audio_separation/Demucs/hdemucs_high_trained.safetensors b/models/audio_separation/Demucs/hdemucs_high_trained.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..996407a23b4ff3f11afa3152bd0ed8861ff6db3b --- /dev/null +++ b/models/audio_separation/Demucs/hdemucs_high_trained.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:26abf3f05f56f7b3618af9448e1aa0580a3ac6ef3c357840be838385b045aab5 +size 334608048 diff --git a/models/audio_separation/Demucs/hdemucs_mmi.safetensors b/models/audio_separation/Demucs/hdemucs_mmi.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8601b148fcc43a296175bf565d6e18cbb6c2680a --- /dev/null +++ b/models/audio_separation/Demucs/hdemucs_mmi.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:92453f0e3e7b08498fd570f2545aed1f92594b934c47127946012ca61bd78a9a +size 167321448 diff --git a/models/audio_separation/Demucs/htdemucs.safetensors b/models/audio_separation/Demucs/htdemucs.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..45434964db670bc6dc1d497f7ec8d8c4ae49719d --- /dev/null +++ b/models/audio_separation/Demucs/htdemucs.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8193504cdfb3943adaf039b8acb524a46e87ebf232c383ac7a32c80a6578423e +size 84030696 diff --git a/models/audio_separation/Demucs/htdemucs_6s.safetensors b/models/audio_separation/Demucs/htdemucs_6s.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..cba302c2e839e48d5b430545075f59a87f917f12 --- /dev/null +++ b/models/audio_separation/Demucs/htdemucs_6s.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f56fe666f0bbf3a645764856ed90f8c5dd8cf4430b1d9649b94e29ec45aa9057 +size 54890960 diff --git a/models/audio_separation/Demucs/htdemucs_ft.safetensors b/models/audio_separation/Demucs/htdemucs_ft.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b4842f8ec5c376ddd103adfe37f61814782dfd62 --- /dev/null +++ b/models/audio_separation/Demucs/htdemucs_ft.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:255c2650d26537ce4887c9c4cf08c6d4896fad2fecc0b78dc5b875b117bcc575 +size 336125008 diff --git a/models/audio_separation/Demucs/mdx.safetensors b/models/audio_separation/Demucs/mdx.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..05903268a9756fb8198182b225fceb46c137ef69 --- /dev/null +++ b/models/audio_separation/Demucs/mdx.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8461e332bcf1848c62b2c6b643584c73a250443d2ce2b4b7abca76fda0681438 +size 691079424 diff --git a/models/audio_separation/Demucs/mdx_extra.safetensors b/models/audio_separation/Demucs/mdx_extra.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c4085cc53a91e00419533549546d0c9fd5d5a018 --- /dev/null +++ b/models/audio_separation/Demucs/mdx_extra.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7999a47e6cb9a2c040bdd2e960db776feee50ddbb27ce29f609de57b31b8dc42 +size 669270384 diff --git a/models/audio_separation/Demucs/repro_mdx_a.safetensors b/models/audio_separation/Demucs/repro_mdx_a.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..183ff9288c98c7e0e16f8beea109a3b6070d26fb --- /dev/null +++ b/models/audio_separation/Demucs/repro_mdx_a.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bced456ec25ff90e008167e1725eb9dd0e4bd7de3ca38dcf4a0a25b1738d8b9c +size 869720032 diff --git a/models/audio_separation/Demucs/repro_mdx_a_hybrid_only.yaml b/models/audio_separation/Demucs/repro_mdx_a_hybrid_only.yaml new file mode 100644 index 0000000000000000000000000000000000000000..d7420c394c8f8fde29484202887e792e8273312b --- /dev/null +++ b/models/audio_separation/Demucs/repro_mdx_a_hybrid_only.yaml @@ -0,0 +1,2 @@ +models: ['fa0cb7f9', '902315c2', 'fa0cb7f9', '902315c2'] +segment: 44 diff --git a/models/audio_separation/Demucs/repro_mdx_a_time_only.yaml b/models/audio_separation/Demucs/repro_mdx_a_time_only.yaml new file mode 100644 index 0000000000000000000000000000000000000000..eb1f442209cabd38c0470aa5a513e269160a7959 --- /dev/null +++ b/models/audio_separation/Demucs/repro_mdx_a_time_only.yaml @@ -0,0 +1,2 @@ +models: ['9a6b4851', '9a6b4851', '1ef250f1', '1ef250f1'] +segment: 44 diff --git a/models/audio_separation/MDX/Kim_Inst.safetensors b/models/audio_separation/MDX/Kim_Inst.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0c3abf9c7ffb6e2b093df74b1379e26cc3c619f0 --- /dev/null +++ b/models/audio_separation/MDX/Kim_Inst.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:71e7f5c6163cea91d77e761f660e0feeaa1ab7fb7100920c69bc77a19bbe5456 +size 66761008 diff --git a/models/audio_separation/MDX/Kim_Vocal_1.safetensors b/models/audio_separation/MDX/Kim_Vocal_1.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8bf0db73fda6b77d705314845c782d21f9712337 --- /dev/null +++ b/models/audio_separation/MDX/Kim_Vocal_1.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ff825df0a49a4b8142f1cf0fabaab13aaa061e250d84ac85e1daf41a791a037f +size 66761000 diff --git a/models/audio_separation/MDX/Kim_Vocal_2.safetensors b/models/audio_separation/MDX/Kim_Vocal_2.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..9bf73d59225f92c52ad609364b12f4d48bcab785 --- /dev/null +++ b/models/audio_separation/MDX/Kim_Vocal_2.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb9d59cbbaf6a9b71b57a9e611c630dd8e5b48e57ecdc73352cea8b30fd6f1a1 +size 66761000 diff --git a/models/audio_separation/MDX/Reverb_HQ_By_FoxJoy.safetensors b/models/audio_separation/MDX/Reverb_HQ_By_FoxJoy.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f0121764e1e64579b0f4535904f1f43c279f991c --- /dev/null +++ b/models/audio_separation/MDX/Reverb_HQ_By_FoxJoy.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8dbd8f8fb5e6164175266fc3d3d42bbc0229eed0616bb83b901c7752b4d323ea +size 66761016 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_1.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_1.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..41a41dcbe8047d7ac8fc0a1dc732ffc5ed5d6fe7 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_1.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0f2abde76ba6b6b52253ba0e69affb8235a008512fa7cd57737f9f303649e9b6 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_2.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_2.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..6f9568b0b85b1e16f0e5eb12c6b6ca2ab2b0ee75 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_2.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2114e2ff89a53f1148efe495faea8671e0e5c63a82a1ddf28b25d5182965b352 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_3.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_3.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..25a3f7b04ecc58e3e7a7e0dc6ecbfb8fdd389716 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_3.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:08d43c2d4c19a2b0ccbee1c8da5e17bed435ae82bd0e1d6f76f3c61c7f67a023 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_1.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_1.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a049f84040c519a56cef05451f6089198584eaa6 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_1.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:19b204105810f40ed57743ee8d0b7dbe6b0fcde01e7a22da13242cb8064c9e2f +size 66761032 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_2.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_2.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0e248ba8e7dd4859b0e783b2417f7e1b35113408 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_2.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7fb0617b075d252f279618898f110561c98890c3cd07a9a279a2b793afdf44bc +size 66761032 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_3.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_3.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ef8a98c28bfa1eb40083bf62264905c3aba8ed07 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_3.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5757892fb859b8523c6b06be1124b5510c558bbaa19fbe0e1900106c9c1117ed +size 66761032 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_4.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_4.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d5c530ef53d49ad9a273560704267263d9cc7ee3 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_4.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bd51a470d38ec5b6985c3bdd3e8690c106f3e54bea4b24e11016ce2d083a20b +size 59076168 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_5.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_5.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..16e473082eb2980829645550e0429ccf1d66e109 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_HQ_5.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fb1b71a1f8c74f68ee0482f41164708a61f51080c370224b4a0a631a9cb96e95 +size 59076168 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_Main.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_Main.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..a4af8799409724ba32a4f11d78b60a134a1c7f1f --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_Main.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:44c26ed251ae43138f598e626cf89a865d3a7de7b464a5ba6deef67d8adc020d +size 52788544 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Inst_full_292.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Inst_full_292.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..ad23e4efa520ef0e985b6604fa100da56349f071 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Inst_full_292.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8aca788315b4919120dcd1ce00c22c26df6c526aeeacc884d8b3828bb8692753 +size 66761040 diff --git a/models/audio_separation/MDX/UVR-MDX-NET-Voc_FT.safetensors b/models/audio_separation/MDX/UVR-MDX-NET-Voc_FT.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d7b50c3d2a1d6b00f6f42150ec638e6bf88933dd --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET-Voc_FT.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4646ea0628b07cdf4e9c352bd89a5b2a7f902b5eb7d06a5afc59136c58d9704 +size 66761008 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Crowd_HQ_1.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Crowd_HQ_1.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f6cf952a501dfffd5bf33b6e1af04a66b91ff3b5 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Crowd_HQ_1.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aa19a7cdbb7726c1d17f3bad97aa38b4e2683f089666a8158d5fc1c56f13d05a +size 59076160 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Inst_187_beta.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Inst_187_beta.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..564a6ad38aa4c5a3116169be5a3f86d06dafb981 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Inst_187_beta.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5dfad26d3bfd625e91ea6d963dd4afddc368d7419a90545842d9a6a36299c1f2 +size 66761040 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Inst_82_beta.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Inst_82_beta.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..364ace3d9389dd9f4077ed093188c609ba3a1abe --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Inst_82_beta.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4a01254fdecc89b2381c02738ded6a1e9e910660b6d0b1ace723f9af49641fcb +size 66761040 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Inst_90_beta.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Inst_90_beta.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d6546fef98bfcbaf8fbd91bf00bbe6bbc780ed15 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Inst_90_beta.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1145a9ac273bcc5c1a1a69826311b4cf7878d1554e36ff3a1d8a0c3360a74a32 +size 66761040 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Main_340.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Main_340.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..86df3c7155075ee5762a2236812f4b312d041f5d --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Main_340.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5b31fe309f2f338a88527b4a4d9e3b8add341f30b93e409d2edba6318a58eb90 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Main_390.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Main_390.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7b39aa3e6d389aac15bbe295029a5b5f477a5de7 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Main_390.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1ac6a965265dee05837dc0db873d9606bc3033f5eada8338e6636ec4ac66b9dd +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Main_406.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Main_406.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1ad6cfb46236ecde5f210729bc51ef37f5e54573 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Main_406.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef534c3d4baadb6af209449aabe9e81e76d57f9875f4663afd2a4b03e68cf398 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Main_427.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Main_427.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..b2b737bed9666153b1f48645abb90b0dfd437b99 --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Main_427.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e6ce4ad341b03c8631e47ccee3301b810d44b6b8d0e904279afb572c3dd3c782 +size 66761024 diff --git a/models/audio_separation/MDX/UVR-MDX-NET_Main_438.safetensors b/models/audio_separation/MDX/UVR-MDX-NET_Main_438.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fda78d7992d364cedf9cc95cf27e3649f32c9c0e --- /dev/null +++ b/models/audio_separation/MDX/UVR-MDX-NET_Main_438.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed8ed60d6266785db1ba21432ecf96ae8961db0837d1ef735c2f3c2bf48557c3 +size 66761024 diff --git a/models/audio_separation/MDX/UVR_MDXNET_1_9703.safetensors b/models/audio_separation/MDX/UVR_MDXNET_1_9703.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..54e5555a1d1a6325020eeba3f4daf1c153db2497 --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_1_9703.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:96ff976433d9995b484cdcdec7e35dd85131d93326d3c208f28101bc0cb249fa +size 29706120 diff --git a/models/audio_separation/MDX/UVR_MDXNET_2_9682.safetensors b/models/audio_separation/MDX/UVR_MDXNET_2_9682.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..2a2b0334a7b0fbec91501d6b46337005e6fb744b --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_2_9682.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bc0dfdd3e8ec197d9732635b301f0802661ac54ba7995277ee010c6f5906b9e1 +size 29706120 diff --git a/models/audio_separation/MDX/UVR_MDXNET_3_9662.safetensors b/models/audio_separation/MDX/UVR_MDXNET_3_9662.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..206cc5f86c7fdbeb5a6a1144d7e603843944488f --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_3_9662.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d7b491f8af4c47b507465816794c01eedf47af6cc07d5406a703719527be89fc +size 29706120 diff --git a/models/audio_separation/MDX/UVR_MDXNET_9482.safetensors b/models/audio_separation/MDX/UVR_MDXNET_9482.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..d33dcccf6a86ec2c71b259847d783065555c016e --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_9482.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8816049691c98035d2192ba55b533a012fdbdc44faf4dc52bc9eb2a0accb6011 +size 29706112 diff --git a/models/audio_separation/MDX/UVR_MDXNET_KARA.safetensors b/models/audio_separation/MDX/UVR_MDXNET_KARA.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3e9cb880173cde7fa22a72b80e6eeb71b71572b6 --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_KARA.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ac69dece16066f84789761fe99d1201f17b3a757092078cb766b978b5779dd83 +size 29706136 diff --git a/models/audio_separation/MDX/UVR_MDXNET_KARA_2.safetensors b/models/audio_separation/MDX/UVR_MDXNET_KARA_2.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f58e49ea46c1b9979ee947b5e226147ad08ea54d --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_KARA_2.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:44373440598e3216a6963d8ca8eaa4ab481051982c1216dcf10f4ded81995a50 +size 52788560 diff --git a/models/audio_separation/MDX/UVR_MDXNET_Main.safetensors b/models/audio_separation/MDX/UVR_MDXNET_Main.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..57bac3cbc6e502b72548cf1aad3c86c921f00e52 --- /dev/null +++ b/models/audio_separation/MDX/UVR_MDXNET_Main.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5759ab6b385668dc61c0b5b32ad2604135b93d49831811fc71fb71cb179d3ed7 +size 66761008 diff --git a/models/audio_separation/MDX/kuielab_a_bass.safetensors b/models/audio_separation/MDX/kuielab_a_bass.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7c0c643ac4c4b686a3a691eedf2d4c4c43f5c542 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_a_bass.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5607dfde927f9946febc8a33b8b2ea63dbdf2a58eac8e0ed9d416a4319bb70a +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_a_drums.safetensors b/models/audio_separation/MDX/kuielab_a_drums.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..f1b2c9000ae6418bcd1545612ef950a33939aa88 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_a_drums.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d9b79d5e693dee35b3f6c6c629f4f8c019f90ab0ddb4305f841bee85fa0bafe9 +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_a_other.safetensors b/models/audio_separation/MDX/kuielab_a_other.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..7c1826f17332f09169a0684d348700088f483805 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_a_other.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e897cf279fa8f08681c1995635f812f1a1bfcb9251f1bf42c9a24f050a9e259 +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_a_vocals.safetensors b/models/audio_separation/MDX/kuielab_a_vocals.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..0a0c846f35bf9ac1b78d385f94e7bdf725dabcee --- /dev/null +++ b/models/audio_separation/MDX/kuielab_a_vocals.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5de1ae905f753230dd516537dce8638572f98dcc9eb2f656bd1f555d85766e92 +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_b_bass.safetensors b/models/audio_separation/MDX/kuielab_b_bass.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3e56fa76ce93b21df10e7aba09f644b99df5a3f7 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_b_bass.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8485e4ccb77c69d8b8faee6032133d58e5c02538712f95d7e3bf9a02291839fa +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_b_drums.safetensors b/models/audio_separation/MDX/kuielab_b_drums.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1f27f29c709fddaff39aee88fa03e5734688bef7 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_b_drums.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2b4acfa2ed8931ec558479a600eca4246942921357dda5bc12c08b7c84aa62df +size 21932680 diff --git a/models/audio_separation/MDX/kuielab_b_other.safetensors b/models/audio_separation/MDX/kuielab_b_other.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..40d6ba543c2e38a109f8d3179e4096d095ba3721 --- /dev/null +++ b/models/audio_separation/MDX/kuielab_b_other.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:915c120c3a8a292c11afd519caf160bc05286b8235a483c0b49e71284d711c27 +size 29706112 diff --git a/models/audio_separation/MDX/kuielab_b_vocals.safetensors b/models/audio_separation/MDX/kuielab_b_vocals.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..84255a15b6e88f8afb8128ef3ef39169b4ecb0ab --- /dev/null +++ b/models/audio_separation/MDX/kuielab_b_vocals.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc8d057a7db0fe96f29836beea6a01391596a8021ef5d4c96cac99512690c293 +size 29706112 diff --git a/models/audio_separation/README.md b/models/audio_separation/README.md new file mode 100644 index 0000000000000000000000000000000000000000..a5e3bf0dbe4a6678d52e217a7cba87d02c910789 --- /dev/null +++ b/models/audio_separation/README.md @@ -0,0 +1,25 @@ +--- +license: mit +base_model: +- Politrees/UVR_resources +tags: +- audio +- separation +- UVR +- vocals +- instruments +- MDX +- MDX-Net +- safetensors +- Demucs +--- + +MDX-Net models from UVR in safetensors format. + +These are all the ones I found in HuggingFace, all claimed to be MIT License. + +Converted to PyTorch for the ComfyUI-AudioSeparator nodes + +--- + +I also added the Demucs v2/4 from Meta converted to safetensors \ No newline at end of file diff --git a/models/audio_separation/source.txt b/models/audio_separation/source.txt new file mode 100644 index 0000000000000000000000000000000000000000..12230d652ab9832ae69e2adaaf03553cf9cbc6bf --- /dev/null +++ b/models/audio_separation/source.txt @@ -0,0 +1 @@ +https://huggingface.co/set-soft/audio_separation \ No newline at end of file