SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.06992 · 2024

Scaling up masked audio encoder learning for general audio classification

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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 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/2406.06992.json · badge: /api/ran/2406.06992.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
richermans/dasheng canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
average_models Ran
ran; our contract draft was wrong, not the code
richermans/dasheng/dasheng/train/utils.py
code served (permissive licence) · get_code("66190d878f7327ac")
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 dict. 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 tempfile
import os

torch.manual_seed(0)

# Create temporary directory for model files
temp_dir = tempfile.mkdtemp()

# Create sample model state dicts and save them
model_files = []

# First model file - with 'model' key structure
state_dict_1 = {
    'model': {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
    },
    'optimizer_state': torch.tensor([1.0, 2.0]),
    'epoch': 1
}
file_1 = os.path.join(temp_dir, 'model_1.pt')
torch.save(state_dict_1, file_1)
model_files.append(file_1)

# Second model file - with 'model' key structure
state_dict_2 = {
    'model': {
        'layer1.weight': torch.randn(4, 3),
        'layer1.bias': torch.randn(4),
    },
    'optimizer_state': torch.tensor([3.0, 4.0]),
    'epoch': 2
}
file_2 = os.path.join(temp_dir, 'model_2.pt')
torch.save(state_dict_2, file_2)
model_files.append(file_2)

# Third model file - without 'model' key (old structure)
state_dict_3 = {
    'layer1.weight': torch.randn(4, 3),
    'layer1.bias': torch.randn(4),
}
file_3 = os.path.join(temp_dir, 'model_3.pt')
torch.save(state_dict_3, file_3)
model_files.append(file_3)

ARGS = {
    'models': model_files
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
convert_decibels_to_amplitude_ratio Ran
ran, honoured the contract we drafted
richermans/dasheng/dasheng/train/audiowebdataset.py
code served (permissive licence) · get_code("6d9e85bfb7ba71c6")
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 numpy as np

torch.manual_seed(0)

# The function convert_decibels_to_amplitude_ratio takes a decibels parameter
# and converts it to an amplitude ratio using the formula: 10 ** (decibels / 20)
# The parameter can be a scalar, numpy array, torch tensor, or any numeric type

# Create a small test case with various input types
decibels_value = 20.0

ARGS = {
    "decibels": decibels_value
}
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 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
proxy_read Ran
ran; our contract draft was wrong, not the code
richermans/dasheng/dasheng/prepare/wavlist_to_tar.py
code served (permissive licence) · get_code("99aea9cc30c39227")
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 dict. 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 json
from pathlib import Path

# Create a temporary file with some binary content
temp_dir = tempfile.mkdtemp()
temp_file = Path(temp_dir) / "test.data.bin"
temp_file.write_bytes(b"binary content here")

# Create the data dict with a filename column
data = {
    "filename_column": str(temp_file),
    "metadata": "some metadata",
    "count": 42
}

ARGS = {
    "data": data,
    "filename_column": "filename_column"
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 3 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.

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

Despite progress in audio classification, a generalization gap remains between speech and other sound domains, such as environmental sounds and music. Models trained for speech tasks often fail to perform well on environmental or musical audio tasks, and vice versa. While self-supervised (SSL) audio representations offer an alternative, there has been limited exploration of scaling both model and dataset sizes for SSL-based general audio classification. We introduce Dasheng, a simple SSL audio encoder, based on the efficient masked autoencoder framework. Trained with 1.2 billion parameters on 272,356 hours of diverse audio, Dasheng obtains significant performance gains on the HEAR benchmark. It outperforms previous works on CREMA-D, LibriCount, Speech Commands, VoxLingua, and competes well in music and environment classification. Dasheng features inherently contain rich speech, music, and environmental information, as shown in nearest-neighbor classification experiments. Code is available https://github.com/richermans/dasheng/.

For agents

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

get_harvested_code_for_paper("2406.06992")
get_code_for_paper("2406.06992")
have("2406.06992")

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

curl https://syntology.ai/api/ran/2406.06992.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.06992.svg)](https://syntology.ai/paper/2406.06992)

Connect an agent — have() is free.