SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1806.03185 · 2018

Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation

arXiv · PDF · Open in the Atlas

Code that ran

12 of 19 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 11 have no contract verdict; 1 failure of our instrument; 7 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/1806.03185.json · badge: /api/ran/1806.03185.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
f90/Wave-U-Net canonical (classifier label)
classifier basis: census
3 ran · 6 no recorded run
f90/Wave-U-Net-Pytorch community (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
haoxiangsnr/Wave-U-Net-for-Speech-Enhancement community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
satvik-venkatesh/Wave-U-net-TF2 community (classifier label)
classifier basis not recorded on the link
1 ran · 1 no recorded run
MattSegal/speech-enhancement extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
build_sinc_filter Ran
ran on a synthesized input
f90/Wave-U-Net-Pytorch/model/resample.py
code served (permissive licence) · get_code("78f6f864d89c2756")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 1, dtype family float, shape (11,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# The function build_sinc_filter takes two parameters:
# - kernel_size: int, must be odd (kernel_size % 2 == 1)
# - cutoff: float, frequency cutoff parameter (0 < cutoff < 1 typically)

ARGS = {
    "kernel_size": 11,  # Odd number as required by assertion
    "cutoff": 0.3,      # Valid cutoff frequency between 0 and 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (11,) — shape (11,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
centre_crop Ran
ran on a synthesized input
f90/Wave-U-Net-Pytorch/model/crop.py
code served (permissive licence) · get_code("2803fec1aab36834")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 3, dtype family float, shape (2, 3, 6) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create input tensor x with shape (batch=2, channels=3, length=10)
# The function crops along the last dimension
x = torch.randn(2, 3, 10)

# Create target tensor with shape (batch=2, channels=3, length=6)
# This will trigger center cropping since 10 - 6 = 4, which is even
target = torch.randn(2, 3, 6)

ARGS = {
    "x": x,
    "target": target,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 3, dtype family float, shape (2, 3, 6) — shape (2, 3, 6) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_loss Ran
ran on a synthesized input
f90/Wave-U-Net-Pytorch/model/utils.py
code served (permissive licence) · get_code("950864364ec0b8d2")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a minimal model that satisfies the function's usage
class SeparableModel(nn.Module):
    def __init__(self, separate=True):
        super().__init__()
        self.separate = separate
        self.instruments = ['drums', 'bass', 'vocals']
        self.linear = nn.Linear(4, 5)
    
    def forward(self, inputs, inst=None):
        if self.separate:
            # When called with inst, return dict with that instrument
            output = self.linear(inputs)
            return {inst: output}
        else:
            # When called without inst, return dict with all instruments
            output = self.linear(inputs)
            return {
                'drums': output,
                'bass': output,
                'vocals': output
            }

class NonSeparableModel(nn.Module):
    def __init__(self, separate=False):
        super().__init__()
        self.separate = separate
        self.instruments = ['drums', 'bass', 'vocals']
        self.linear = nn.Linear(4, 5)
    
    def forward(self, inputs, inst=None):
        output = self.linear(inputs)
        return {
            'drums': output,
            'bass': output,
            'vocals': output
        }

# Test with separate model
model_separate = SeparableModel(separate=True)

# Input and target tensors
batch_size = 2
input_dim = 4
output_dim = 5

inputs = torch.randn(batch_size, input_dim)
targets = {
    'drums': torch.randn(batch_size, output_dim),
    'bass': torch.randn(batch_size, output_dim),
    'vocals': torch.randn(batch_size, output_dim)
}

# Loss criterion
criterion = nn.L1Loss()

ARGS = {
    'model': model_separate,
    'inputs': inputs,
    'targets': targets,
    'criterion': criterion,
    'compute_grad': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
getNumParams Ran
ran on a synthesized input
f90/Wave-U-Net/Utils.py
code served (permissive licence) · get_code("404b97b66e233e0d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int64, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np
import tensorflow as tf

# Create small TensorFlow tensors with distinct shapes
tensor1 = tf.constant(np.random.randn(2, 4), dtype=tf.float32)  # shape (2, 4) = 8 params
tensor2 = tf.constant(np.random.randn(3, 5), dtype=tf.float32)  # shape (3, 5) = 15 params
tensor3 = tf.constant(np.random.randn(2), dtype=tf.float32)     # shape (2,) = 2 params

tensors = [tensor1, tensor2, tensor3]

ARGS = {
    "tensors": tensors
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int64, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_lr Ran
ran on a synthesized input
f90/Wave-U-Net-Pytorch/utils.py
code served (permissive licence) · get_code("b293995d32808cdd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
from torch.optim import SGD

torch.manual_seed(0)

# Create a simple model
model = nn.Linear(4, 3)

# Create an optimizer
optim = SGD(model.parameters(), lr=0.01)

# Construct arguments
ARGS = {
    'optim': optim
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_padding Ran
ran on a synthesized input
satvik-venkatesh/Wave-U-net-TF2/wave_u_net.py
code served (permissive licence) · get_code("d8372ce707c3095c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 2, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

ARGS = {
    'shape': [2, 128, 1],
    'num_layers': 12,
    'filter_size': 15,
    'input_filter_size': 15,
    'output_filter_size': 1,
    'merge_filter_size': 5,
    'num_channels': 1,
    'context': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family int — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
initialize_config Ran
ran on a synthesized input
haoxiangsnr/Wave-U-Net-for-Speech-Enhancement/util/utils.py
code served (permissive licence) · get_code("aafe9da12cdcd105")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned StandInModel. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import importlib
import sys
from types import ModuleType

# Create a stand-in module with a simple class
stand_in_module = ModuleType("stand_in_model")

class StandInModel:
    def __init__(self, hidden_dim=64, num_layers=2):
        self.hidden_dim = hidden_dim
        self.num_layers = num_layers
    
    def __call__(self, x):
        return x

stand_in_module.MyModel = StandInModel

# Register the module so importlib can find it
sys.modules["stand_in_model"] = stand_in_module

# Define ARGS
ARGS = {
    "module_cfg": {
        "module": "stand_in_model",
        "main": "MyModel",
        "args": {
            "hidden_dim": 128,
            "num_layers": 3
        }
    },
    "pass_args": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type StandInModel — a StandInModel output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StandInModel output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_checkpoint Ran
ran on a synthesized input
haoxiangsnr/Wave-U-Net-for-Speech-Enhancement/util/utils.py
code served (permissive licence) · get_code("8fdba4f9ae28cbe8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import torch
import tempfile

torch.manual_seed(0)

# Create temporary directory for checkpoint files
tmpdir = tempfile.mkdtemp()

# Create a .pth checkpoint file
pth_checkpoint_path = os.path.join(tmpdir, 'model.pth')
pth_data = {'weights': torch.randn(2, 4)}
torch.save(pth_data, pth_checkpoint_path)

# Create a .tar checkpoint file with epoch info and model data
tar_checkpoint_path = os.path.join(tmpdir, 'model.tar')
tar_data = {
    'epoch': 42,
    'model': {'weights': torch.randn(3, 4)}
}
torch.save(tar_data, tar_checkpoint_path)

device = 'cpu'

ARGS = {
    'checkpoint_path': pth_checkpoint_path,
    'device': device
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_model Ran
ran on a synthesized input
f90/Wave-U-Net-Pytorch/model/utils.py
code served (permissive licence) · get_code("abf55a235181f2e0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import tempfile
import os

torch.manual_seed(0)

# Create a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create model instance
model = SimpleModel()

# Create optimizer
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

# Create a checkpoint file
checkpoint = {
    'model_state_dict': model.state_dict(),
    'optimizer_state_dict': optimizer.state_dict(),
    'step': 42
}

# Save checkpoint to a temporary file
with tempfile.NamedTemporaryFile(delete=False, suffix='.pt') as f:
    checkpoint_path = f.name
    torch.save(checkpoint, checkpoint_path)

# Define arguments
ARGS = {
    'model': model,
    'optimizer': optimizer,
    'path': checkpoint_path,
    'cuda': False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
subsample_chunk_random Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
MattSegal/speech-enhancement/src/datasets/speech/noisy_librispeech/librispeech_dataset.py
pointer only (licence: NONE) · get_code("19db27f0a2d1b55c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, MattSegal/speech-enhancement when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

take_all_snippets Ran
ran on a synthesized input
f90/Wave-U-Net/Datasets.py
code served (permissive licence) · get_code("773e07dfbc4df2db")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned _TensorSliceDataset. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf

# Create a sample dict with data and length
sample = {
    'feature_a': tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0], [7.0, 8.0, 9.0], [10.0, 11.0, 12.0], [13.0, 14.0, 15.0]], dtype=tf.float32),
    'feature_b': tf.constant([[16.0, 17.0, 18.0], [19.0, 20.0, 21.0], [22.0, 23.0, 24.0], [25.0, 26.0, 27.0], [28.0, 29.0, 30.0]], dtype=tf.float32),
    'length': tf.constant(5, dtype=tf.int64)
}

keys = ['feature_a', 'feature_b']
input_shape = (2, 3)
output_shape = (1,)

ARGS = {
    'sample': sample,
    'keys': keys,
    'input_shape': input_shape,
    'output_shape': output_shape
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type _TensorSliceDataset — a _TensorSliceDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _TensorSliceDataset output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
take_snippets_at_pos Ran
ran on a synthesized input
f90/Wave-U-Net/Datasets.py
code served (permissive licence) · get_code("b8955c9d5cd1a2ed")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned _TensorSliceDataset. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create sample data - a dictionary with keys mapped to tensors
# The function accesses sample[key][pos:pos + input_shape[0], :], so we need
# tensors with at least 2 dimensions where first dimension is for indexing
sample = {
    'feature_a': tf.constant(np.random.randn(20, 3).astype(np.float32)),
    'feature_b': tf.constant(np.random.randn(20, 3).astype(np.float32))
}

# Keys to process
keys = ['feature_a', 'feature_b']

# Start positions for each sample
start_pos = tf.constant([0, 5], dtype=tf.int32)

# Input shape: [sequence_length, feature_dim]
input_shape = [4, 3]

# Number of samples being extracted
num_samples = 2

ARGS = {
    'sample': sample,
    'keys': keys,
    'start_pos': start_pos,
    'input_shape': input_shape,
    'num_samples': num_samples
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type _TensorSliceDataset — a _TensorSliceDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _TensorSliceDataset output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
compute_mean_metrics Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Evaluate.py
code served (permissive licence) · get_code("7ef7210907a0d9be")
How it ran

no invocation recorded.

crop_and_concat Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Utils.py
code served (permissive licence) · get_code("f3054ff75a9faf20")
How it ran

no invocation recorded.

getTrainableVariables Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Utils.py
code served (permissive licence) · get_code("e119d722ad3969f8")
How it ran

no invocation recorded.

independent_outputs Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Models/OutputLayer.py
code served (permissive licence) · get_code("7fd1fa630f0e11bd")
How it ran

no invocation recorded.

learned_interpolation_layer Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Models/InterpolationLayer.py
code served (permissive licence) · get_code("0af32c8aa2b13465")
How it ran

no invocation recorded.

take_random_snippets Not yet run
unverified (harvested, no recorded run)
f90/Wave-U-Net/Datasets.py
code served (permissive licence) · get_code("6950f34674f12037")
How it ran

no invocation recorded.

wave_u_net Not yet run
unverified (harvested, no recorded run)
satvik-venkatesh/Wave-U-net-TF2/wave_u_net.py
code served (permissive licence) · get_code("8efc19a229bb0bc7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 16 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
build_sinc_filter 1 other paper
1 later
2208.11428 (2022-08)
Automatic music mixing with deep learning and out-of-domain data
getNumParams 1 other paper
1 dated earlier
1711.00048 (2017-11)
Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice Extraction
getTrainableVariables 1 other paper
1 dated earlier
1711.00048 (2017-11)
Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice Extraction

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Models for audio source separation usually operate on the magnitude spectrum, which ignores phase information and makes separation performance dependant on hyper-parameters for the spectral front-end. Therefore, we investigate end-to-end source separation in the time-domain, which allows modelling phase information and avoids fixed spectral transformations. Due to high sampling rates for audio, employing a long temporal input context on the sample level is difficult, but required for high quality separation results because of long-range temporal correlations. In this context, we propose the Wave-U-Net, an adaptation of the U-Net to the one-dimensional time domain, which repeatedly resamples feature maps to compute and combine features at different time scales. We introduce further architectural improvements, including an output layer that enforces source additivity, an upsampling technique and a context-aware prediction framework to reduce output artifacts. Experiments for singing voice separation indicate that our architecture yields a performance comparable to a state-of-the-art spectrogram-based U-Net architecture, given the same data. Finally, we reveal a problem with outliers in the currently used SDR evaluation metrics and suggest reporting rank-based statistics to alleviate this problem.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1806.03185")
get_code_for_paper("1806.03185")
have("1806.03185")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/1806.03185.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/1806.03185.svg)](https://syntology.ai/paper/1806.03185)

Connect an agent — have() is free.