SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.18636 · 2024

A Diffusion-Based Generative Equalizer for Music Restoration

arXiv · PDF · Open in the Atlas

Code that ran

8 of 9 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, 0 violated it and 0 have no contract verdict; 5 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/2403.18636.json · badge: /api/ran/2403.18636.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
eloimoliner/babe2 canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
FunctionStatusWhere it lives
apply_filter Ran
ran; our contract draft was wrong, not the code
eloimoliner/babe2/utils/blind_bwe_utils.py
code served (permissive licence) · get_code("fb82da2f3594f544")
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, 128) 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)

# Construct minimal valid arguments for apply_filter
x = torch.randn(2, 128)  # batch_size=2, time_samples=128
NFFT = 64  # FFT size
H = torch.randn(2, 33)  # batch_size=2, freq_bins=33 (NFFT//2 + 1 for one-sided FFT)

ARGS = {
    "x": x,
    "H": H,
    "NFFT": NFFT,
}
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 2, dtype family float, shape (2, 128) — shape (2, 128) 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_time Ran
ran; our contract draft was wrong, not the code
eloimoliner/babe2/utils/dnnlib/util.py
code served (permissive licence) · get_code("674eca7b9e1b6439")
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) 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 numpy as np
from typing import Union

# Test with various time values
ARGS = {
    "seconds": 3661.7  # 1 hour, 1 minute, 1.7 seconds - tests rounding and all format branches
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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_time_brief Ran
ran; our contract draft was wrong, not the code
eloimoliner/babe2/utils/dnnlib/util.py
code served (permissive licence) · get_code("32af001972a0699f")
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) 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 numpy as np
from typing import Union

# Test with various time values
ARGS = {
    "seconds": 45.7  # A float value that will be rounded to 46 seconds
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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
get_FIR_high_pass Ran
ran, honoured the contract we drafted
eloimoliner/babe2/utils/bandwidth_extension.py
code served (permissive licence) · get_code("ef3bbf76b4e5fc56")
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 (1, 1, 31) 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 scipy.signal

torch.manual_seed(0)

ARGS = {
    'order': 32,
    'fc': 0.2,
    'beta': 5.0,
    'sr': 44100.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
scipy, 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 (1, 1, 31) — shape (1, 1, 31) 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_FIR_lowpass Ran
ran, honoured the contract we drafted
eloimoliner/babe2/utils/bandwidth_extension.py
code served (permissive licence) · get_code("738fe3789883750b")
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 (1, 1, 32) 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 scipy.signal

torch.manual_seed(0)

ARGS = {
    "order": 32,
    "fc": 0.25,
    "beta": 5.0,
    "sr": 16000
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
scipy, 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 (1, 1, 32) — shape (1, 1, 32) 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
prepare_filter Ran
ran, honoured the contract we drafted
eloimoliner/babe2/utils/bandwidth_extension.py
code served (permissive licence) · get_code("16e4f29b6b9f674b")
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 (1, 1, 33) 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 math

torch.manual_seed(0)

class FilterConfig:
    def __init__(self, filter_type, order, fc, beta=None, ripple=None, Q=None, resample_fs=None, decimate_factor=None):
        self.type = filter_type
        self.order = order
        self.fc = fc
        self.beta = beta
        self.ripple = ripple
        self.biquad = type('obj', (object,), {'Q': Q})() if Q is not None else None
        self.resample = type('obj', (object,), {'fs': resample_fs})() if resample_fs is not None else None

class BandwidthExtensionConfig:
    def __init__(self, filter_config):
        self.filter = filter_config
        self.decimate = type('obj', (object,), {'factor': 2})()

class TesterConfig:
    def __init__(self, bandwidth_extension_config):
        self.bandwidth_extension = bandwidth_extension_config

class ArgsConfig:
    def __init__(self, tester_config):
        self.tester = tester_config

# Create different filter configurations for testing
def create_args_for_filter_type(filter_type):
    if filter_type == 'firwin':
        filter_cfg = FilterConfig('firwin', order=33, fc=0.3, beta=8.6)
    elif filter_type == 'firwin_hpf':
        filter_cfg = FilterConfig('firwin_hpf', order=33, fc=0.3, beta=8.6)
    elif filter_type == 'cheby1':
        filter_cfg = FilterConfig('cheby1', order=4, fc=0.3, ripple=0.1)
    elif filter_type == 'biquad':
        filter_cfg = FilterConfig('biquad', order=1, fc=1000.0, Q=0.707)
    elif filter_type == 'resample':
        filter_cfg = FilterConfig('resample', order=1, fc=0.5, resample_fs=8000)
    elif filter_type == 'decimate':
        filter_cfg = FilterConfig('decimate', order=1, fc=0.5)
    else:
        filter_cfg = FilterConfig('firwin', order=33, fc=0.3, beta=8.6)
    
    bw_ext_cfg = BandwidthExtensionConfig(filter_cfg)
    tester_cfg = TesterConfig(bw_ext_cfg)
    args = ArgsConfig(tester_cfg)
    return args

ARGS = {
    'args': create_args_for_filter_type('firwin'),
    'sample_rate': 16000
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
scipy, 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 (1, 1, 33) — shape (1, 1, 33) 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
smooth_LTAS Ran
ran; our contract draft was wrong, not the code
eloimoliner/babe2/utils/blind_bwe_utils.py
code served (permissive licence) · get_code("f9f6a2abd0b08bbe")
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 (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)

# LTAS: a 1D tensor representing Long-Term Average Spectrum
# The function clones it, indexes it, and performs operations on it
LTAS = torch.randn(4)

# f: a 1D tensor of frequencies
# The function iterates from 1 to len(f), indexes it, and passes elements to gauss_f
# It needs to support indexing and len()
f = torch.tensor([100.0, 200.0, 300.0, 400.0])

# Noct: number of octaves for smoothing
# The function checks if Noct > 0 and uses it in division
Noct = 1

ARGS = {
    "LTAS": LTAS,
    "f": f,
    "Noct": Noct,
}
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 (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
weight_init Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
eloimoliner/babe2/networks/cqtdiff.py
code served (permissive licence) · get_code("d41a4250066bce93")
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 (4, 3) 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)

ARGS = {
    'shape': (4, 3),
    'mode': 'xavier_uniform',
    'fan_in': 10,
    'fan_out': 8,
}
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 (4, 3) — shape (4, 3) 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
ask_yes_no Not yet run
unverified (harvested, no recorded run)
eloimoliner/babe2/utils/dnnlib/util.py
code served (permissive licence) · get_code("9d31d2c4cd16bb2d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 4 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
format_time 73 other papers
26 dated earlier, 4 the same month, 42 later, 1 undated
2004.00049 (2020-04)
In-Domain GAN Inversion for Real Image Editing
2004.11660 (2020-04)
Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learning
2108.00946 (2021-08)
StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generators
and 70 more not named here
ask_yes_no 69 other papers
26 dated earlier, 4 the same month, 38 later, 1 undated
2004.00049 (2020-04)
In-Domain GAN Inversion for Real Image Editing
2004.11660 (2020-04)
Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learning
2108.00946 (2021-08)
StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generators
and 66 more not named here
format_time_brief 68 other papers
22 dated earlier, 4 the same month, 41 later, 1 undated
2108.00946 (2021-08)
StyleGAN-NADA: CLIP-Guided Domain Adaptation of Image Generators
2202.00273 (2022-02)
StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasets
2206.07695 (2022-06)
VoxGRAF: Fast 3D-Aware Image Synthesis with Sparse Voxel Grids
and 65 more not named here
weight_init 49 other papers
13 dated earlier, 3 the same month, 33 later · 1 only as an assembly brick
2303.10137 (2023-03)
A Recipe for Watermarking Diffusion Models
2305.15266 (2023-05)
Diffusion-Based Audio Inpainting
2305.16269 (2023-05)
UDPM: Upsampling Diffusion Probabilistic Models
and 46 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

This paper presents a novel approach to audio restoration, focusing on the enhancement of low-quality music recordings, and in particular historical ones. Building upon a previous algorithm called BABE, or Blind Audio Bandwidth Extension, we introduce BABE-2, which presents a series of improvements. This research broadens the concept of bandwidth extension to \emph{generative equalization}, a novel task that, to the best of our knowledge, has not been explicitly addressed in previous studies. BABE-2 is built around an optimization algorithm utilizing priors from diffusion models, which are trained or fine-tuned using a curated set of high-quality music tracks. The algorithm simultaneously performs two critical tasks: estimation of the filter degradation magnitude response and hallucination of the restored audio. The proposed method is objectively evaluated on historical piano recordings, showing an enhancement over the prior version. The method yields similarly impressive results in rejuvenating the works of renowned vocalists Enrico Caruso and Nellie Melba. This research represents an advancement in the practical restoration of historical music.

For agents

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

get_harvested_code_for_paper("2403.18636")
get_code_for_paper("2403.18636")
have("2403.18636")

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

curl https://syntology.ai/api/ran/2403.18636.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.18636.svg)](https://syntology.ai/paper/2403.18636)

Connect an agent — have() is free.