SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.08100 · 2020

Conformer: Convolution-augmented Transformer for Speech Recognition

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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, 3 violated it and 1 has no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-21; 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/2005.08100.json · badge: /api/ran/2005.08100.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
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
lucidrains/conformer reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
msalhab96/Conformer community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
TeaPoly/Conformer-Athena extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
jaketae/conformer community (classifier label)
classifier basis not recorded on the link
1 ran · 0 no recorded run
TensorSpeech/TensorFlowASR community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
calc_same_padding Ran
ran, violated the contract we drafted
lucidrains/conformer/conformer/conformer.py
code served (permissive licence) · get_code("a41cb31b42e919b1")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned tuple, ndim 1, dtype family int; 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.

ARGS = {
    'kernel_size': 5
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 tuple, ndim 1, dtype family int — 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
check_token Ran
ran; our contract draft was wrong, not the code
msalhab96/Conformer/tokenizer.py
code served (permissive licence) · get_code("d759ddcfb8763ee8")
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
from typing import Callable

torch.manual_seed(0)

# Create a mock object with _token_to_id attribute
class MockTokenizer:
    def __init__(self):
        self._token_to_id = {
            "hello": 1,
            "world": 2,
            "test": 3
        }

# Create a mock function to be decorated
def mock_function(obj, token):
    """Mock function that would be decorated"""
    return -1  # Return a default value if token not found

# The check_token decorator takes a token string as parameter
ARGS = {
    "token": "hello"
}
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
default Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from ThomasMrY/VCT
pointer only; the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("60fff7c3c400d7ff")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

exists Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from ThomasMrY/VCT
pointer only; the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("aa5486a3650902d8")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

get_speakers Ran
ran; our contract draft was wrong, not the code
TeaPoly/Conformer-Athena/examples/asr/timit/local/prepare_data.py
code served (permissive licence) · get_code("2835e46864d2299b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned list, 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 os
import tempfile
import shutil

# Create a temporary directory structure to simulate the expected file layout
temp_dir = tempfile.mkdtemp()
local_dir = os.path.join(temp_dir, 'examples', 'asr', 'timit', 'local')
os.makedirs(local_dir, exist_ok=True)

# Create a mock speaker list file for 'dev' set
dev_spk_file = os.path.join(local_dir, 'dev_spk.list')
with open(dev_spk_file, 'w') as f:
    f.write('speaker1\n')
    f.write('speaker2\n')
    f.write('speaker3\n')

# Create a mock speaker list file for 'test' set
test_spk_file = os.path.join(local_dir, 'test_spk.list')
with open(test_spk_file, 'w') as f:
    f.write('speakerA\n')
    f.write('speakerB\n')

# Change to the temporary directory so relative paths work
original_cwd = os.getcwd()
os.chdir(temp_dir)

# Define the arguments
ARGS = {
    'set_name': 'dev'
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 list, ndim 1, dtype family float — a list 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
skew Ran
ran on a synthesized input
jaketae/conformer/conformer/attention.py
code served (permissive licence) · get_code("746f1b671084d3ba")
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 4, dtype family float, shape (2, 3, 4, 5) 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 torch.nn.functional as F

torch.manual_seed(0)

# QEr is a 4D tensor with shape (batch_size, num_heads, num_rows, num_cols)
# The function pads it, reshapes it, and extracts a slice
# Using distinct sizes: batch=2, heads=3, rows=4, cols=5
QEr = torch.randn(2, 3, 4, 5)

ARGS = {
    "QEr": QEr
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
deserialize Not yet run
unverified (harvested, no recorded run)
TensorSpeech/TensorFlowASR/tensorflow_asr/callbacks.py
code served (permissive licence) · get_code("efd738a0bc3cc757")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 5 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 0 the same month, 4 later) and 2 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
exists 486 other papers
18 dated earlier, 466 later, 2 undated · 78 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1804.04637 (2018-04)
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
and 483 more not named here
default 163 other papers
6 dated earlier, 156 later, 1 undated · 27 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1810.00825 (2018-10)
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
1907.01470 (2019-07)
Augmenting Self-attention with Persistent Memory
and 160 more not named here
calc_same_padding 4 other papers
4 later
2202.01855 (2022-02)
Self-Supervised Learning with Random-Projection Quantizer for Speech Recognition
2203.15149 (2022-03)
CMGAN: Conformer-based Metric GAN for Speech Enhancement
2310.01889 (2023-10)
Ring Attention with Blockwise Transformers for Near-Infinite Context
and 1 more not named here
check_token 1 other paper
1 later
2006.04664 (2020-06)
MultiSpeech: Multi-Speaker Text to Speech with Transformer
deserialize 1 other paper
1 dated earlier
1703.09452 (2017-03)
SEGAN: Speech Enhancement Generative Adversarial Network

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recently Transformer and Convolution neural network (CNN) based models have shown promising results in Automatic Speech Recognition (ASR), outperforming Recurrent neural networks (RNNs). Transformer models are good at capturing content-based global interactions, while CNNs exploit local features effectively. In this work, we achieve the best of both worlds by studying how to combine convolution neural networks and transformers to model both local and global dependencies of an audio sequence in a parameter-efficient way. To this regard, we propose the convolution-augmented transformer for speech recognition, named Conformer. Conformer significantly outperforms the previous Transformer and CNN based models achieving state-of-the-art accuracies. On the widely used LibriSpeech benchmark, our model achieves WER of 2.1%/4.3% without using a language model and 1.9%/3.9% with an external language model on test/testother. We also observe competitive performance of 2.7%/6.3% with a small model of only 10M parameters.

For agents

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

get_harvested_code_for_paper("2005.08100")
get_code_for_paper("2005.08100")
have("2005.08100")

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

curl https://syntology.ai/api/ran/2005.08100.json

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

[![Syntology run record](https://syntology.ai/api/ran/2005.08100.svg)](https://syntology.ai/paper/2005.08100)

Connect an agent — have() is free.