SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.06875 · CVPR · 2022

Visual Acoustic Matching

Kristen Grauman, Changan Chen, Ruohan Gao, Paul Calamia

arXiv · PDF · Open in the Atlas

Code that ran

8 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 2 failures of our instrument; 4 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/2202.06875.json · badge: /api/ran/2202.06875.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
see2sound/see2sound community
Papers with Code archive
8 ran · 4 no recorded run
FunctionStatusWhere it lives
ReLUSine Ran
ran on a synthesized input
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("b8488faf5efce39c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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 torch.nn as nn

class ReLUSine(nn.Module):

    def __init__(self):
        super().__init__()

    def forward(self, input):
        a = torch.sin(30 * input)
        b = nn.ReLU(inplace=False)(input)
        return a + b


INIT_ARGS = {}

ARGS = {
    'input': torch.randn(2, 4)
}

if __name__ == '__main__':
    model = ReLUSine(**INIT_ARGS)
    output = model(**ARGS)
    print(output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Sine Ran
ran on a synthesized input · constructed an object; no result was computed
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("198244478a93e51e")
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 an instance of its own class, Sine: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

ARGS = {
    'freq': 2.0,
    'gain': 1.5
}
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 Sine — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Sine 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
create_dummy_class Ran
ran on a synthesized input
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("bcb61d59810720e3")
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 dummy_class. 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)

class dummy_class:
    pass

def create_dummy_class():
    dummy_class_item = dummy_class()
    dummy_class_item.input_blocks = []
    dummy_class_item.input_block_connecters_in = []
    for i in range(12):
        dummy_class_item.input_blocks.append(range(2))
        dummy_class_item.input_block_connecters_in.append(None)
    dummy_class_item.middle_block = range(3)
    dummy_class_item.output_blocks = []
    dummy_class_item.output_block_connecters_in = []
    for i in range(12):
        dummy_class_item.output_blocks.append(range(2))
        dummy_class_item.output_block_connecters_in.append(None)
    return dummy_class_item

ARGS = {}
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 dummy_class — a dummy_class output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dummy_class 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
dummy_class Ran
ran on a synthesized input · constructed an object; no result was computed
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("70e02a8bc79113a6")
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 an instance of its own class, dummy_class: the run constructed the object and did not call it, so no result was computed. 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)

# dummy_class has no __init__ parameters, so ARGS should be an empty dict
ARGS = {}
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 dummy_class — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a dummy_class 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_unit Ran
ran on a synthesized input · constructed an object; no result was computed
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("2cae2bfdc5c08f10")
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 an instance of its own class, get_unit: the run constructed the object and did not call it, so no result was computed. 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)

# The get_unit class is a singleton that registers and retrieves neural network units.
# Looking at the __init__ method, it takes no parameters.
# The __call__ method is used to parse unit names and return either the unit class
# or a functools.partial with parsed arguments.

# Since get_unit.__init__ takes no parameters, there are no parameters to map in ARGS.
# However, let's verify by understanding what get_unit does:
# - It's decorated with @singleton
# - __init__ takes only self
# - It registers various units (conv, bn, relu, sine, relusine, etc.)
# - __call__ parses string descriptions and returns unit classes or partials

# The singleton decorator wraps the class and the getinstance function takes *args and **kwargs
# but get_unit.__init__ only takes self, so no additional arguments are needed.

ARGS = {}
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 get_unit — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a get_unit 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
singleton Ran
ran; our contract draft was wrong, not the code
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("a57e939b71fd9a3d")
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 function. 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)

# The singleton function takes a class_ parameter
# It creates instances dictionary and returns getinstance function
# The function stores instances of the class in a dictionary keyed by class_

class SimpleClass:
    def __init__(self, value=42):
        self.value = value

ARGS = {
    "class_": SimpleClass
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
str2value Ran
ran, honoured the contract we drafted
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("bdc6bdc2bcb6c992")
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 int, 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 torch
import numpy as np

def str2value(v):
    v = v.strip()
    try:
        return int(v)
    except:
        pass
    try:
        return float(v)
    except:
        pass
    if v in ('True', 'true'):
        return True
    elif v in ('False', 'false'):
        return False
    else:
        return v

ARGS = {
    'v': '  42  '
}
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 int, 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
timestep_embedding Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("7e83c65f054fb972")
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 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 math
from einops import repeat

torch.manual_seed(0)

# Create timesteps tensor: 1-D tensor of N indices
timesteps = torch.tensor([0.0, 1.0], dtype=torch.float32)

# Create dimension parameter
dim = 4

# Create max_period parameter
max_period = 10000

# Create repeat_only parameter
repeat_only = False

ARGS = {
    "timesteps": timesteps,
    "dim": dim,
    "max_period": max_period,
    "repeat_only": repeat_only,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
einops, 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
UNetModelCoDi Not yet run
unverified (harvested, no recorded run)
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("53da3ce697277473")
How it ran

no invocation recorded.

get_model Not yet run
unverified (harvested, no recorded run)
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("66b6a436203b0aad")
How it ran

no invocation recorded.

preprocess_model_args Not yet run
unverified (harvested, no recorded run)
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("67a656fb4beb59ca")
How it ran

no invocation recorded.

register Not yet run
unverified (harvested, no recorded run)
see2sound/see2sound/see2sound/codi/models/latent_diffusion/diffusion_unet.py
code served (permissive licence) · get_code("c73767b33cb559fa")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 later) and 10 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
timestep_embedding
brick of this paper's assembly: helper
25 other papers
25 later · 8 only as an assembly brick
2301.05225 (2023-01)
Domain Expansion of Image Generators
2302.07685 (2023-02)
Video Probabilistic Diffusion Models in Projected Latent Space
2305.16283 (2023-05)
CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graph Diffusion
and 22 more not named here
singleton
brick of this paper's assembly: helper
2 other papers
2 later
Xu_Versatile_Diffusion_Text_Images_and_Variations_All_in_One_Diffusion_ICCV_2023_paper (2023 (year only))
Versatile Diffusion: Text, Images and Variations All in One Diffusion Model
2305.11846 (2023-05)
Any-to-Any Generation via Composable Diffusion

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

We introduce the visual acoustic matching task, in which an audio clip is transformed to sound like it was recorded in a target environment. Given an image of the target environment and a waveform for the source audio, the goal is to re-synthesize the audio to match the target room acoustics as suggested by its visible geometry and materials. To address this novel task, we propose a cross-modal transformer model that uses audio-visual attention to inject visual properties into the audio and generate realistic audio output. In addition, we devise a self-supervised training objective that can learn acoustic matching from in-the-wild Web videos, despite their lack of acoustically mismatched audio. We demonstrate that our approach successfully translates human speech to a variety of real-world environments depicted in images, outperforming both traditional acoustic matching and more heavily supervised baselines.

For agents

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

get_harvested_code_for_paper("2202.06875")
get_code_for_paper("2202.06875")
have("2202.06875")

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

curl https://syntology.ai/api/ran/2202.06875.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.06875.svg)](https://syntology.ai/paper/2202.06875)

Connect an agent — have() is free.