SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.14307 · 2023

Auto-AVSR: Audio-Visual Speech Recognition with Automatic Labels

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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 3 have no contract verdict; 0 failures of our instrument; 3 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/2303.14307.json · badge: /api/ran/2303.14307.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
mpc001/auto_avsr canonical (classifier label)
classifier basis not recorded on the link
3 ran · 3 no recorded run
FunctionStatusWhere it lives
collate_pad Ran
ran on a synthesized input
mpc001/auto_avsr/datamodule/data_module.py
code served (permissive licence) · get_code("0a672d203cd8007f")
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

torch.manual_seed(0)

# Build sample data matching the expected structure
# batch is a list of dicts, each dict has keys like 'input', 'target', etc.
# Each value is either None or a tensor

sample1_input = torch.randn(5, 4)  # length 5, feature dim 4
sample1_target = torch.randint(0, 10, (5,))  # length 5, scalar targets

sample2_input = torch.randn(3, 4)  # length 3, feature dim 4
sample2_target = torch.randint(0, 10, (3,))  # length 3, scalar targets

batch = [
    {
        'input': sample1_input,
        'target': sample1_target,
    },
    {
        'input': sample2_input,
        'target': sample2_target,
    }
]

ARGS = {
    'batch': batch
}
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
end_detect Ran
ran on a synthesized input
mpc001/auto_avsr/espnet/nets/e2e_asr_common.py
code served (permissive licence) · get_code("41d108cb26723328")
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 bool, 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 numpy as np

# Construct ended_hyps: a list of hypothesis dicts
# Each hypothesis dict needs:
# - 'score': a float (used for sorting and comparison)
# - 'yseq': a sequence (used for len() to get hypothesis length)
ended_hyps = [
    {'score': -5.0, 'yseq': [1, 2, 3, 4]},
    {'score': -6.0, 'yseq': [1, 2, 3]},
    {'score': -7.0, 'yseq': [1, 2]},
    {'score': -8.0, 'yseq': [1, 2, 3, 4, 5]},
    {'score': -9.0, 'yseq': [1]},
]

# i: current index/step
i = 5

# M: number of hypotheses to check (default 3)
M = 3

# D_end: threshold for end detection (default log(1 * exp(-10)))
D_end = np.log(1 * np.exp(-10))

ARGS = {
    'ended_hyps': ended_hyps,
    'i': i,
    'M': M,
    'D_end': D_end,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
pad Ran
ran on a synthesized input
mpc001/auto_avsr/datamodule/data_module.py
code served (permissive licence) · get_code("593d603cc0b0a93f")
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 tuple. 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)

# Create sample tensors with different shapes
# The function handles 1D, 2D, and 4D tensors
sample_1d = torch.randn(5)  # 1D tensor: length 5
sample_2d = torch.randn(7, 3)  # 2D tensor: length 7, feature dim 3
sample_4d = torch.randn(4, 2, 2, 2)  # 4D tensor: length 4, then 2x2x2

# Test case 1: 1D tensors
samples_1d = [
    torch.randn(3),
    torch.randn(5),
]

# Test case 2: 2D tensors
samples_2d = [
    torch.randn(4, 2),
    torch.randn(6, 2),
]

# Test case 3: 4D tensors
samples_4d = [
    torch.randn(3, 2, 2, 2),
    torch.randn(5, 2, 2, 2),
]

ARGS = {
    "samples": samples_2d,
    "pad_val": 0.0,
}
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 tuple — a tuple output has no array shape of its own; none is 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
average_checkpoints Not yet run
unverified (harvested, no recorded run)
mpc001/auto_avsr/average_checkpoints.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from umbertocappellazzo/llama-avsr, is pointer only (licence: NONE) · get_code("3ae2eab75af094e1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from umbertocappellazzo/llama-avsr, and GitHub detected no licence in that repository 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. No invocation recorded.

ensemble Not yet run
unverified (harvested, no recorded run)
mpc001/auto_avsr/average_checkpoints.py
code served (permissive licence) · get_code("eddc218242d34add")
How it ran

no invocation recorded.

load_video Not yet run
unverified (harvested, no recorded run)
mpc001/auto_avsr/datamodule/av_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from umbertocappellazzo/llama-avsr, is pointer only (licence: NONE) · get_code("e3ffd0b6ffbdf70f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from umbertocappellazzo/llama-avsr, and GitHub detected no licence in that repository 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. No invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 4 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
average_checkpoints 1 other paper
1 later
2409.12319 (2024-09)
Large Language Models are Strong Audio-Visual Speech Recognition Learners
load_video 1 other paper
1 later
2409.12319 (2024-09)
Large Language Models are Strong Audio-Visual Speech Recognition Learners

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

Audio-visual speech recognition has received a lot of attention due to its robustness against acoustic noise. Recently, the performance of automatic, visual, and audio-visual speech recognition (ASR, VSR, and AV-ASR, respectively) has been substantially improved, mainly due to the use of larger models and training sets. However, accurate labelling of datasets is time-consuming and expensive. Hence, in this work, we investigate the use of automatically-generated transcriptions of unlabelled datasets to increase the training set size. For this purpose, we use publicly-available pre-trained ASR models to automatically transcribe unlabelled datasets such as AVSpeech and VoxCeleb2. Then, we train ASR, VSR and AV-ASR models on the augmented training set, which consists of the LRS2 and LRS3 datasets as well as the additional automatically-transcribed data. We demonstrate that increasing the size of the training set, a recent trend in the literature, leads to reduced WER despite using noisy transcriptions. The proposed model achieves new state-of-the-art performance on AV-ASR on LRS2 and LRS3. In particular, it achieves a WER of 0.9% on LRS3, a relative improvement of 30% over the current state-of-the-art approach, and outperforms methods that have been trained on non-publicly available datasets with 26 times more training data.

For agents

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

get_harvested_code_for_paper("2303.14307")
get_code_for_paper("2303.14307")
have("2303.14307")

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

curl https://syntology.ai/api/ran/2303.14307.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.14307.svg)](https://syntology.ai/paper/2303.14307)

Connect an agent — have() is free.