SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2409.13382 · 2024

Audio Codec Augmentation for Robust Collaborative Watermarking of Speech Synthesis

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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, 2 violated it and 0 have no contract verdict; 2 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/2409.13382.json · badge: /api/ran/2409.13382.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
ljuvela/collaborative-watermarking-with-codecs canonical (classifier label)
classifier basis: census
4 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_det_curve Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ljuvela/collaborative-watermarking-with-codecs/src/collaborative_watermarking/metrics.py
code served (permissive licence) · get_code("047149960a2fd5d6")
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 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

# Create small but distinct numpy arrays for target and nontarget scores
target_scores = np.array([0.8, 0.7, 0.9, 0.75])
nontarget_scores = np.array([0.3, 0.4, 0.2, 0.5, 0.1])

ARGS = {
    'target_scores': target_scores,
    'nontarget_scores': nontarget_scores,
}
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 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
compute_eer Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ljuvela/collaborative-watermarking-with-codecs/src/collaborative_watermarking/metrics.py
code served (permissive licence) · get_code("6c660ac679abd4e2")
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 numpy as np

# Create sample target scores (authentic speaker trials)
target_scores = np.array([0.85, 0.92, 0.88, 0.91, 0.87])

# Create sample nontarget scores (impostor/spoofed trials)
nontarget_scores = np.array([0.25, 0.35, 0.30, 0.40, 0.28, 0.32])

ARGS = {
    "target_scores": target_scores,
    "nontarget_scores": nontarget_scores,
}
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 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
dynamic_range_compression Ran
ran, violated the contract we drafted
ljuvela/collaborative-watermarking-with-codecs/src/collaborative_watermarking/meldataset.py
code served (permissive licence) · get_code("be031b15a988a556")
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 (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 numpy as np

# Construct a small numpy array as input
x = np.array([0.00001, 0.0001, 0.001, 0.01, 0.1, 1.0, 10.0, 100.0])

# Define arguments dict with all parameters
ARGS = {
    "x": x,
    "C": 1,
    "clip_val": 1e-05
}
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 (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
dynamic_range_decompression Ran
ran, violated the contract we drafted
ljuvela/collaborative-watermarking-with-codecs/src/collaborative_watermarking/meldataset.py
code served (permissive licence) · get_code("9e30e7d510c9d341")
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 (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 numpy as np

# The function takes two parameters: x and C
# x is expected to be a numpy array (based on np.exp(x) operation)
# C is a scalar value used for division (default is 1)

# Create a small numpy array for x
x = np.array([0.5, 1.0, 1.5, 2.0])

# Create a scalar value for C
C = 2.0

ARGS = {
    "x": x,
    "C": C
}
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 (4,) — shape (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

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 2 the same month, 2 later) and 0 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
dynamic_range_compression 13 other papers
8 dated earlier, 2 the same month, 3 later
2006.04558 (2020-06)
FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
2106.07889 (2021-06)
UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation
2110.12676 (2021-10)
Controllable and Interpretable Singing Voice Decomposition via Assem-VC
and 10 more not named here
dynamic_range_decompression 13 other papers
8 dated earlier, 2 the same month, 3 later
2006.04558 (2020-06)
FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
2106.07889 (2021-06)
UnivNet: A Neural Vocoder with Multi-Resolution Spectrogram Discriminators for High-Fidelity Waveform Generation
2110.12676 (2021-10)
Controllable and Interpretable Singing Voice Decomposition via Assem-VC
and 10 more not named here
compute_det_curve 8 other papers
8 dated earlier
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
6 dated earlier
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

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

Automatic detection of synthetic speech is becoming increasingly important as current synthesis methods are both near indistinguishable from human speech and widely accessible to the public. Audio watermarking and other active disclosure methods of are attracting research activity, as they can complement traditional deepfake defenses based on passive detection. In both active and passive detection, robustness is of major interest. Traditional audio watermarks are particularly susceptible to removal attacks by audio codec application. Most generated speech and audio content released into the wild passes through an audio codec purely as a distribution method. We recently proposed collaborative watermarking as method for making generated speech more easily detectable over a noisy but differentiable transmission channel. This paper extends the channel augmentation to work with non-differentiable traditional audio codecs and neural audio codecs and evaluates transferability and effect of codec bitrate over various configurations. The results show that collaborative watermarking can be reliably augmented by black-box audio codecs using a waveform-domain straight-through-estimator for gradient approximation. Furthermore, that results show that channel augmentation with a neural audio codec transfers well to traditional codecs. Listening tests demonstrate collaborative watermarking incurs negligible perceptual degradation with high bitrate codecs or DAC at 8kbps.

For agents

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

get_harvested_code_for_paper("2409.13382")
get_code_for_paper("2409.13382")
have("2409.13382")

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

curl https://syntology.ai/api/ran/2409.13382.json

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

[![Syntology run record](https://syntology.ai/api/ran/2409.13382.svg)](https://syntology.ai/paper/2409.13382)

Connect an agent — have() is free.