SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.01428 · 2023

Improved DeepFake Detection Using Whisper Features

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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; 1 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 0 failures of our instrument; 0 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/2306.01428.json · badge: /api/ran/2306.01428.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
piotrkawa/deepfake-whisper-features canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_eer Ran
ran on a synthesized input
piotrkawa/deepfake-whisper-features/src/metrics.py
code served (permissive licence) · get_code("671d92010ac58822")
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 numpy as np
from sklearn.metrics import roc_curve

# Set random seed for reproducibility
np.random.seed(0)

# Create synthetic binary classification data
# y: true binary labels
# y_score: predicted scores (higher = more likely positive class)

# Create 10 samples with binary labels
y = np.array([0, 0, 0, 0, 0, 1, 1, 1, 1, 1])

# Create prediction scores that are somewhat discriminative
# Negatives get lower scores, positives get higher scores
y_score = np.array([0.1, 0.2, 0.3, 0.25, 0.15, 0.7, 0.8, 0.75, 0.85, 0.9])

ARGS = {
    "y": y,
    "y_score": y_score,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, sklearn — 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
exact_div Ran
ran, honoured the contract we drafted
piotrkawa/deepfake-whisper-features/src/models/whisper_main.py
pointer only (licence: MIT); the stored copy get_code serves, from qwenlm/qwen-audio, is pointer only (licence: NOASSERTION) · get_code("6e0808fc4828513a")
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 qwenlm/qwen-audio, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

forward_and_loss Ran
ran on a synthesized input
piotrkawa/deepfake-whisper-features/src/trainer.py
code served (permissive licence) · get_code("23ba206b58cf64d6")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Define a simple model that takes batch_x and produces output
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Define a simple criterion (loss function)
criterion = nn.MSELoss()

# Create batch_x: shape (2, 4) - batch size 2, feature dim 4
batch_x = torch.randn(2, 4)

# Create batch_y: shape (2, 3) - batch size 2, target dim 3
batch_y = torch.randn(2, 3)

# Create model instance
model = SimpleModel()

ARGS = {
    'model': model,
    'criterion': criterion,
    'batch_x': batch_x,
    'batch_y': batch_y
}
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 — 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
mel_filters Ran
ran on a synthesized input
piotrkawa/deepfake-whisper-features/src/models/whisper_main.py
code served (permissive licence) · get_code("287787d54e6c2146")
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; 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 numpy as np
import os
from unittest.mock import patch, MagicMock

torch.manual_seed(0)

# Create a mock for the mel_filters.npz file since we can't do file I/O
mock_mel_data = {
    'mel_80': np.random.randn(80, 201).astype(np.float32)
}

# Patch np.load to return our mock data
original_load = np.load

def mock_npz_load(filepath, *args, **kwargs):
    if 'mel_filters.npz' in filepath:
        return MagicMock(__enter__=lambda self: mock_mel_data, __exit__=lambda self, *args: None)
    return original_load(filepath, *args, **kwargs)

# Apply the patch before defining ARGS
np.load = mock_npz_load

ARGS = {
    'device': 'cpu',
    'n_mels': 80
}
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 — no output shape is stated: observe_signature.py could not observe it (call:NameError)
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_or_trim Ran
ran on a synthesized input
piotrkawa/deepfake-whisper-features/src/models/whisper_main.py
code served (permissive licence) · get_code("b7043f3ab423733b")
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, 480000) 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)

SAMPLE_RATE = 16000
CHUNK_LENGTH = 30
N_SAMPLES = CHUNK_LENGTH * SAMPLE_RATE

# Test case 1: torch tensor input
audio_tensor = torch.randn(2, 8000)

# Test case 2: numpy array input
audio_numpy = np.random.randn(2, 4000).astype(np.float32)

ARGS = {
    "array": audio_tensor,
    "length": N_SAMPLES,
    "axis": -1,
}
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, 480000) — shape (2, 480000) 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 4 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
exact_div 6 other papers
1 dated earlier, 5 later
2212.04356 (2022-12)
Robust Speech Recognition via Large-Scale Weak Supervision
2308.06112 (2023-08)
Lip2Vec: Efficient and Robust Visual Speech Recognition via Latent-to-Latent Visual to Audio Representation Mapping
2309.15701 (2023-09)
HyPoradise: An Open Baseline for Generative Speech Recognition with Large Language Models
and 3 more not named here

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

With a recent influx of voice generation methods, the threat introduced by audio DeepFake (DF) is ever-increasing. Several different detection methods have been presented as a countermeasure. Many methods are based on so-called front-ends, which, by transforming the raw audio, emphasize features crucial for assessing the genuineness of the audio sample. Our contribution contains investigating the influence of the state-of-the-art Whisper automatic speech recognition model as a DF detection front-end. We compare various combinations of Whisper and well-established front-ends by training 3 detection models (LCNN, SpecRNet, and MesoNet) on a widely used ASVspoof 2021 DF dataset and later evaluating them on the DF In-The-Wild dataset. We show that using Whisper-based features improves the detection for each model and outperforms recent results on the In-The-Wild dataset by reducing Equal Error Rate by 21%.

For agents

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

get_harvested_code_for_paper("2306.01428")
get_code_for_paper("2306.01428")
have("2306.01428")

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

curl https://syntology.ai/api/ran/2306.01428.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.01428.svg)](https://syntology.ai/paper/2306.01428)

Connect an agent — have() is free.