SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1910.10261 · 2019

QuartzNet: Deep Automatic Speech Recognition with 1D Time-Channel Separable Convolutions

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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, 0 violated it and 7 have no contract verdict; 3 failures of our instrument; 2 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/1910.10261.json · badge: /api/ran/1910.10261.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
oleges1/quartznet-pytorch community (classifier label)
classifier basis not recorded on the link
4 ran · 0 no recorded run
marka17/digit-recognition community (classifier label)
classifier basis not recorded on the link
3 ran · 2 no recorded run
Kirili4ik/QuartzNet-ASR-pytorch reimplementation (classifier label)
classifier basis: classifier_2025
2 ran · 0 no recorded run
isadrtdinov/quartznet reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
conv_bn_act Ran
ran; our contract draft was wrong, not the code
Kirili4ik/QuartzNet-ASR-pytorch/model.py
pointer only (licence: NONE) · get_code("93ea36b1e45e4172")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Kirili4ik/QuartzNet-ASR-pytorch when it was checked. 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.

convert_bytes Ran
ran on a synthesized input
marka17/digit-recognition/gfile.py
code served (permissive licence) · get_code("fa009e15ffeb21b1")
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 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 torch
import numpy as np

def convert_bytes(num):
    """
    this function will convert bytes to MB.... GB... etc
    """
    for x in ['bytes', 'KB', 'MB', 'GB', 'TB']:
        if num < 1024.0:
            return '%3.1f %s' % (num, x)
        num /= 1024.0

ARGS = {
    'num': 2048.0
}
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 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
convert_human_time Ran
ran on a synthesized input
marka17/digit-recognition/gfile.py
code served (permissive licence) · get_code("0cdbba0bc41e103e")
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 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 torch
import numpy as np

# Construct valid arguments for convert_human_time
ARGS = {
    "seconds": 3661.5,  # 1 hour, 1 minute, 1.5 seconds
    "digits": 2
}
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 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_act_dropout_layer Ran
ran on a synthesized input
oleges1/quartznet-pytorch/model/layers.py
code served (permissive licence) · get_code("97c5b22c22d161c6")
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 Sequential. 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
from torch import nn

torch.manual_seed(0)

ARGS = {
    'drop_prob': 0.2,
    'activation': 'relu'
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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_conv_bn_layer Ran
ran on a synthesized input
oleges1/quartznet-pytorch/model/layers.py
code served (permissive licence) · get_code("c13d6372fb3fe4ad")
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 Sequential. 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
from torch import nn

torch.manual_seed(0)

ARGS = {
    'in_channels': 8,
    'out_channels': 16,
    'kernel_size': 5,
    'stride': 1,
    'dilation': 1,
    'padding': 2,
    'bias': False,
    'groups': 2,
    'separable': False,
    'normalization': 'batch',
    'norm_groups': 1,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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_same_padding Ran
ran on a synthesized input
marka17/digit-recognition/src/modules/jasper.py
code served (permissive licence) · get_code("ba297314ad959cc0")
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

torch.manual_seed(0)

# The function get_same_padding takes three integer parameters:
# - kernel_size: size of the kernel
# - stride: stride value
# - dilation: dilation value
# 
# The function validates that stride and dilation cannot both be > 1
# and returns kernel_size // 2 * dilation

ARGS = {
    'kernel_size': 3,
    'stride': 1,
    'dilation': 1
}
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
get_same_padding Ran
ran on a synthesized input
oleges1/quartznet-pytorch/model/utils.py
code served (permissive licence) · get_code("6a909b63465cabc2")
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

torch.manual_seed(0)

ARGS = {
    'kernel_size': 3,
    'stride': 1,
    'dilation': 1
}
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
quartznet Ran
ran; our contract draft was wrong, not the code
isadrtdinov/quartznet/asr/models/quartznet.py
code served (permissive licence) · get_code("5fac3b34256008d7")
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 QuartzNet. 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)

# Define parameters for quartznet function
ARGS = {
    'num_labels': 29,
    'params': {
        'num_mels': 128,
        'num_blocks': 5,
        'num_cells': 5,
        'input_kernel': 33,
        'input_channels': 256,
        'head_kernel': 87,
        'head_channels': 512,
        'block_kernels': (33, 39, 51, 63, 75),
        'block_channels': (256, 256, 256, 512, 512),
        'dropout_rate': 0.25
    }
}
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 QuartzNet — a QuartzNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a QuartzNet 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
remove_from_dict Ran
ran on a synthesized input
oleges1/quartznet-pytorch/utils.py
code served (permissive licence) · get_code("f02f10b701054ea3")
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 numpy as np

torch.manual_seed(0)

# Construct arguments for remove_from_dict
the_dict = {
    'key1': 'value1',
    'key2': 'value2',
    'key3': 'value3',
    'key4': 'value4',
}

keys = ['key1', 'key3']

ARGS = {
    'the_dict': the_dict,
    'keys': keys,
}
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 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
sepconv_bn Ran
ran; our contract draft was wrong, not the code
Kirili4ik/QuartzNet-ASR-pytorch/model.py
pointer only (licence: NONE) · get_code("1a3c3bbfbd73d6d8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, Kirili4ik/QuartzNet-ASR-pytorch when it was checked. 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.

calculate_cer Not yet run
unverified (harvested, no recorded run)
marka17/digit-recognition/src/utils/metrics.py
code served (permissive licence) · get_code("8aa28eb81f15e19e")
How it ran

no invocation recorded.

download_file_from_google_drive Not yet run
unverified (harvested, no recorded run)
marka17/digit-recognition/gfile.py
code served (permissive licence) · get_code("dc1cace2b69b6f80")
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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 11 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
download_file_from_google_drive 5 other papers
2 dated earlier, 3 later
1808.10437 (2018-08)
iCAN: Instance-Centric Attention Network for Human-Object Interaction Detection
1811.08264 (2018-11)
Transferable Interactiveness Knowledge for Human-Object Interaction Detection
Bae_Multi-View_Depth_Estimation_by_Fusing_Single-View_Depth_Probability_With_Multi-View_CVPR_2022_paper (2022 (year only))
Multi-View Depth Estimation by Fusing Single-View Depth Probability with Multi-View Geometry
and 2 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

We propose a new end-to-end neural acoustic model for automatic speech recognition. The model is composed of multiple blocks with residual connections between them. Each block consists of one or more modules with 1D time-channel separable convolutional layers, batch normalization, and ReLU layers. It is trained with CTC loss. The proposed network achieves near state-of-the-art accuracy on LibriSpeech and Wall Street Journal, while having fewer parameters than all competing models. We also demonstrate that this model can be effectively fine-tuned on new datasets.

For agents

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

get_harvested_code_for_paper("1910.10261")
get_code_for_paper("1910.10261")
have("1910.10261")

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

curl https://syntology.ai/api/ran/1910.10261.json

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

[![Syntology run record](https://syntology.ai/api/ran/1910.10261.svg)](https://syntology.ai/paper/1910.10261)

Connect an agent — have() is free.