SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2309.07525 · 2023

SingFake: Singing Voice Deepfake Detection

arXiv · PDF · Open in the Atlas

Code that ran

15 of 16 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; 3 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 10 failures of our instrument; 1 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/2309.07525.json · badge: /api/ran/2309.07525.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
yongyizang/SingFake canonical (classifier label)
classifier basis: census
15 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_det_curve Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yongyizang/SingFake/models/feat+resnet/eval_metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from asvspoof-challenge/2021, is pointer only (licence: NONE) · get_code("3a889d3debbcead8")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from asvspoof-challenge/2021, and GitHub detected no licence in that repository 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.

compute_eer Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yongyizang/SingFake/models/feat+resnet/eval_metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from asvspoof-challenge/2021, is pointer only (licence: NONE) · get_code("047cd2d292cdc66d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from asvspoof-challenge/2021, and GitHub detected no licence in that repository 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.

cosine_annealing Ran
ran, honoured the contract we drafted
yongyizang/SingFake/models/AASIST/utils.py
code served (permissive licence) · get_code("0e4e062545eb78ac")
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 float64, ndim 0, 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 numpy as np

ARGS = {
    "step": 5,
    "total_steps": 100,
    "lr_max": 0.1,
    "lr_min": 0.001
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
delta Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/models/feat+resnet/audio_feature_extraction.py
code served (permissive licence) · get_code("58ec4646ad814adf")
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, 5, 4) 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.functional as torch_nn_func

torch.manual_seed(0)

# The function `delta` takes a single parameter `x` which should be a tensor
# of shape (batch, Length, dim)
# It performs a delta/difference operation along the Length dimension

# Create a small tensor with distinct dimensions:
# batch=2, Length=5, dim=4
x = torch.randn(2, 5, 4)

ARGS = {
    'x': x
}
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, 5, 4) — shape (2, 5, 4) 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
format_codec Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/dataset/simulate_codec.py
code served (permissive licence) · get_code("055280b0e9fae51f")
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 str. 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 torch
import numpy as np

torch.manual_seed(0)

# The function format_codec takes a single parameter 'format' which is a string
# It checks the value against 'mp3', 'ogg', 'adts' and returns codec strings or None
# We need to provide a valid string argument

ARGS = {
    'format': 'mp3'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
format_suffix Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/dataset/simulate_codec.py
code served (permissive licence) · get_code("b966dbdba0741d13")
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 str. 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 torch
import numpy as np

ARGS = {
    'format': 'mp3'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
genSpoof_list Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/models/AASIST/data_utils.py
code served (permissive licence) · get_code("54a60bd39fb10f2d")
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. 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 tempfile
import os

# Create a temporary metadata file with the expected format
temp_dir = tempfile.mkdtemp()
meta_file = os.path.join(temp_dir, "meta.txt")

# Write sample metadata in the expected format: col1 key col3 col4 label
meta_content = """speaker_1 utt_001 session_1 protocol_1 bonafide
speaker_2 utt_002 session_2 protocol_2 spoof
speaker_3 utt_003 session_1 protocol_1 bonafide
speaker_4 utt_004 session_2 protocol_2 spoof
"""

with open(meta_file, 'w') as f:
    f.write(meta_content)

ARGS = {
    'dir_meta': meta_file,
    'is_train': True,
    'is_eval': False,
}
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 tuple — a tuple output has no array shape of its own; none is 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
keras_decay Ran
ran, honoured the contract we drafted
yongyizang/SingFake/models/AASIST/utils.py
code served (permissive licence) · get_code("461ce4011175cbad")
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

torch.manual_seed(0)

# The function keras_decay takes:
# - step: an integer or numeric value representing the training step
# - decay: a float representing the decay rate (default 0.0001)
# 
# The function computes: 1.0 / (1.0 + decay * step)
# It performs simple arithmetic operations on these parameters.

ARGS = {
    "step": 100,
    "decay": 0.0001,
}
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
label_2_float Ran
ran, honoured the contract we drafted
yongyizang/SingFake/models/feat+resnet/audio_feature_extraction.py
code served (permissive licence) · get_code("7453b30a6a37676b")
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 2, dtype family float, shape (2, 4) 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 numpy as np

torch.manual_seed(0)

# The function takes:
# x: a tensor of integer values (long or int), any shape
# bits: number of bits (int)

# Create a small tensor with integer values
x = torch.tensor([[0, 1, 2, 3], [4, 5, 6, 7]], dtype=torch.long)

# Set bits to a reasonable value
bits = 8

ARGS = {
    "x": x,
    "bits": bits
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
obtain_asv_error_rates Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/models/feat+resnet/eval_metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from xjchengit/singgraph, is pointer only (licence: NONE) · get_code("4a9fc5f1de58c879")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from xjchengit/singgraph, and GitHub detected no licence in that repository 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.

pad Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yongyizang/SingFake/models/AASIST/data_utils.py
code served (permissive licence) · get_code("43d18bf97d3ba573")
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 (20,) 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

# Examine the function: pad(x, max_len=64600)
# - x: expects a numpy array with .shape[0] attribute
# - max_len: optional int parameter (default 64600)
# The function uses:
#   - x.shape[0] to get first dimension
#   - x[:max_len] for slicing
#   - np.tile() to repeat x
#   - Array indexing operations

# Create a small numpy array for x
x = np.array([1.0, 2.0, 3.0, 4.0, 5.0])

# Create ARGS dict with all parameters
ARGS = {
    "x": x,
    "max_len": 20
}
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 (20,) — shape (20,) 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
pad_random Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
yongyizang/SingFake/models/AASIST/data_utils.py
code served (permissive licence) · get_code("b4f9b637a765b790")
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 (64600,) 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

np.random.seed(0)

# Create a small numpy array as test input
x = np.random.randn(1000)

# Define the maximum length parameter
max_len = 64600

# Construct ARGS dict with all parameters
ARGS = {
    "x": x,
    "max_len": max_len
}
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 (64600,) — shape (64600,) 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
repeat_padding_Tensor Ran
ran; our contract draft was wrong, not the code
yongyizang/SingFake/models/feat+resnet/dataset.py
code served (permissive licence) · get_code("14738aac076461dd")
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, 7, 4) 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 numpy as np

torch.manual_seed(0)

# The function takes a spec tensor and ref_len integer
# spec is accessed with .shape[1] and .repeat(1, mul, 1) operations
# This suggests spec is a 3D tensor with shape (batch, time, features)

spec = torch.randn(2, 3, 4)  # (batch=2, time=3, features=4)
ref_len = 7  # A reference length to pad/repeat to

ARGS = {
    "spec": spec,
    "ref_len": ref_len
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, 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, 7, 4) — shape (2, 7, 4) 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
str_to_bool Ran
ran, violated the contract we drafted
yongyizang/SingFake/models/AASIST/utils.py
code served (permissive licence) · get_code("bc2ace1bc5b21488")
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 bool, 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

# str_to_bool only takes a single parameter 'val' which should be a string
ARGS = {
    'val': 'yes'
}
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 bool, 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
trimf Ran
ran on a synthesized input
yongyizang/SingFake/models/feat+resnet/audio_feature_extraction.py
code served (permissive licence) · get_code("0ddbd12af0260f4b")
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 1, dtype family float, shape (8,) 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 numpy as np

torch.manual_seed(0)

# Create a small tensor for x
x = torch.linspace(0, 10, 8, dtype=torch.float32)

# Define params as [a, b, c] where a <= b <= c
params = [2.0, 5.0, 8.0]

ARGS = {
    'x': x,
    'params': params
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 1, dtype family float, shape (8,) — shape (8,) 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
repeat_padding_RawTensor Not yet run
unverified (harvested, no recorded run)
yongyizang/SingFake/models/feat+resnet/dataset.py
code served (permissive licence) · get_code("d03c780aa3716d38")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 1 the same month, 4 later) and 5 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
cosine_annealing 9 other papers
6 dated earlier, 1 the same month, 1 later, 1 same year, order not determinable
1901.09960 (2019-01)
Using Pre-Training Can Improve Model Robustness and Uncertainty
2003.09711 (2020-03)
Robust Out-of-distribution Detection for Neural Networks
2108.11941 (2021-08)
Semantically Coherent Out-of-Distribution Detection
and 6 more not named here
compute_det_curve 8 other papers
5 dated earlier, 3 later
1904.05441 (2019-04)
ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
2011.01108 (2020-11)
End-to-end anti-spoofing with RawNet2
2104.01320 (2021-04)
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
and 5 more not named here
compute_eer 6 other papers
3 dated earlier, 3 later
2011.01108 (2020-11)
End-to-end anti-spoofing with RawNet2
2104.01320 (2021-04)
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
2210.02437 (2022-10)
ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild
and 3 more not named here
obtain_asv_error_rates 5 other papers
4 dated earlier, 1 later
1904.05441 (2019-04)
ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
2011.01108 (2020-11)
End-to-end anti-spoofing with RawNet2
2104.01320 (2021-04)
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
and 2 more not named here
pad 2 other papers
2 dated earlier
2110.01200 (2021-10)
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
2111.04433 (2021-11)
RawBoost: A Raw Data Boosting and Augmentation Method applied to Automatic Speaker Verification Anti-Spoofing
genSpoof_list 1 other paper
1 dated earlier
2110.01200 (2021-10)
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
keras_decay 1 other paper
1 dated earlier
2110.01200 (2021-10)
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
label_2_float 1 other paper
1 dated earlier
2104.01320 (2021-04)
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
pad_random 1 other paper
1 dated earlier
2110.01200 (2021-10)
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks
repeat_padding_Tensor 1 other paper
1 dated earlier
2104.01320 (2021-04)
An Empirical Study on Channel Effects for Synthetic Voice Spoofing Countermeasure Systems
str_to_bool 1 other paper
1 dated earlier
2110.01200 (2021-10)
AASIST: Audio Anti-Spoofing using Integrated Spectro-Temporal Graph Attention Networks

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

The rise of singing voice synthesis presents critical challenges to artists and industry stakeholders over unauthorized voice usage. Unlike synthesized speech, synthesized singing voices are typically released in songs containing strong background music that may hide synthesis artifacts. Additionally, singing voices present different acoustic and linguistic characteristics from speech utterances. These unique properties make singing voice deepfake detection a relevant but significantly different problem from synthetic speech detection. In this work, we propose the singing voice deepfake detection task. We first present SingFake, the first curated in-the-wild dataset consisting of 28.93 hours of bonafide and 29.40 hours of deepfake song clips in five languages from 40 singers. We provide a train/validation/test split where the test sets include various scenarios. We then use SingFake to evaluate four state-of-the-art speech countermeasure systems trained on speech utterances. We find these systems lag significantly behind their performance on speech test data. When trained on SingFake, either using separated vocal tracks or song mixtures, these systems show substantial improvement. However, our evaluations also identify challenges associated with unseen singers, communication codecs, languages, and musical contexts, calling for dedicated research into singing voice deepfake detection. The SingFake dataset and related resources are available at https://www.singfake.org/.

For agents

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

get_harvested_code_for_paper("2309.07525")
get_code_for_paper("2309.07525")
have("2309.07525")

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

curl https://syntology.ai/api/ran/2309.07525.json

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

[![Syntology run record](https://syntology.ai/api/ran/2309.07525.svg)](https://syntology.ai/paper/2309.07525)

Connect an agent — have() is free.