SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.03563 · 2024

Learning Video Temporal Dynamics with Cross-Modal Attention for Robust Audio-Visual Speech Recognition

arXiv · PDF · Open in the Atlas

Code that ran

12 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 10 failures of our instrument; 1 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/2407.03563.json · badge: /api/ran/2407.03563.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
sungnyun/avsr-temporal-dynamics canonical (classifier label)
classifier basis: census
12 ran · 1 no recorded run
FunctionStatusWhere it lives
Embedding Ran
ran; our contract draft was wrong, not the code · constructed an object; no result was computed
sungnyun/avsr-temporal-dynamics/avhubert/hubert_asr.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from instance-wise-ordered-transformer/iot, is pointer only (licence: NOASSERTION) · get_code("96cbb5e9ca5b6be0")
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 instance-wise-ordered-transformer/iot, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

Linear Ran
ran; our contract draft was wrong, not the code · constructed an object; no result was computed
sungnyun/avsr-temporal-dynamics/avhubert/hubert_asr.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from instance-wise-ordered-transformer/iot, is pointer only (licence: NOASSERTION) · get_code("8cd8cb0d1e9e63e4")
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 instance-wise-ordered-transformer/iot, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

apply_half Ran
ran, honoured the contract we drafted
sungnyun/avsr-temporal-dynamics/avhubert/infer_s2s.py
code served (permissive licence) · get_code("357d3a4f0be679d9")
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

torch.manual_seed(0)

# The function apply_half takes a single parameter 't' which should be a tensor
# It checks if t.dtype is torch.float32 and converts to torch.half if so
# Otherwise returns t unchanged

# Create a small float32 tensor to test the conversion
t = torch.randn(2, 4, dtype=torch.float32)

ARGS = {
    "t": t
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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, 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
compute_mask_indices Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sungnyun/avsr-temporal-dynamics/avhubert/utils.py
code served (permissive licence) · get_code("581de97f5d5aa97a")
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
import numpy as np

torch.manual_seed(0)
np.random.seed(0)

# Define the ARGS dictionary with valid arguments for compute_mask_indices
ARGS = {
    'shape': (2, 10),  # batch_size=2, timesteps=10
    'padding_mask': torch.tensor([[False, False, False, False, False, False, False, False, False, False],
                                   [False, False, False, False, True, True, True, True, True, True]]),
    'mask_prob': 0.15,
    'mask_length': 3,
    'mask_type': 'static',
    'mask_other': 0.0,
    'min_masks': 1,
    'no_overlap': False,
    'min_space': 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 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from amerch/CIFAR100-Training, is pointer only (licence: NONE) · get_code("583f9780bdd00a45")
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 amerch/CIFAR100-Training, 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. A run is recorded.

downsample_basic_block Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from umbertocappellazzo/llama-avsr, is pointer only (licence: NONE) · get_code("d50e3a8f0b0f9a21")
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. A run is recorded.

downsample_basic_block_v2 Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from umbertocappellazzo/llama-avsr, is pointer only (licence: NONE) · get_code("546a770034e200d1")
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. A run is recorded.

find_runs Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sungnyun/avsr-temporal-dynamics/avhubert/utils.py
code served (permissive licence) · get_code("db4a33e22cd9f644")
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, ndim 2, 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.

import numpy as np

# Test case 1: Simple array with runs
x = np.array([1, 1, 2, 2, 2, 3, 1, 1])

ARGS = {
    'x': x
}
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 tuple, ndim 2, 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
get_symbols_to_strip_from_output Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/infer_s2s.py
code served (permissive licence) · get_code("bbe1a698e337d4d4")
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 set. 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

class MockGenerator:
    def __init__(self, include_symbols_attr=False):
        self.eos = 2
        self.pad = 0
        if include_symbols_attr:
            self.symbols_to_strip_from_output = {1, 2, 3}

# Test case 1: generator with symbols_to_strip_from_output attribute
generator_with_attr = MockGenerator(include_symbols_attr=True)

ARGS = {
    'generator': generator_with_attr
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
load_label Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/hubert_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yxduir/LLM-SRT, is pointer only (licence: NONE) · get_code("113a523619225ba0")
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 yxduir/LLM-SRT, 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. A run is recorded.

load_label_offset Ran
ran; our contract draft was wrong, not the code
sungnyun/avsr-temporal-dynamics/avhubert/hubert_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yxduir/LLM-SRT, is pointer only (licence: NONE) · get_code("3e9450b6ac2fcddd")
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 yxduir/LLM-SRT, 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. A run is recorded.

load_video Ran
ran, honoured the contract we drafted
sungnyun/avsr-temporal-dynamics/avhubert/utils.py
code served (permissive licence) · get_code("d5a7afb95212031e")
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 ndarray, ndim 3, dtype family int, shape (5, 240, 320) 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 cv2
import numpy as np
import tempfile
import os

# Create a temporary video file for testing
def create_test_video(path, num_frames=5, width=320, height=240):
    """Create a simple test video file."""
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(path, fourcc, 30.0, (width, height))
    
    for i in range(num_frames):
        # Create a simple frame with some variation
        frame = np.full((height, width, 3), fill_value=(i * 50) % 256, dtype=np.uint8)
        out.write(frame)
    
    out.release()

# Create a temporary directory and video file
temp_dir = tempfile.mkdtemp()
video_path = os.path.join(temp_dir, "test_video.mp4")
create_test_video(video_path)

ARGS = {
    "path": video_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (5, 240, 320) — shape (5, 240, 320) 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
load_audio_visual Not yet run
unverified (harvested, no recorded run)
sungnyun/avsr-temporal-dynamics/avhubert/hubert_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yxduir/LLM-SRT, is pointer only (licence: NONE) · get_code("f0641584d0fe0fb5")
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 yxduir/LLM-SRT, 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 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 13 are also present in other papers whose code we hold (13 with at least one paper dated earlier, 2 the same month, 9 later) and 0 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
conv3x3 686 other papers
603 dated earlier, 14 the same month, 52 later, 17 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
Linear 58 other papers
51 dated earlier, 1 the same month, 5 later, 1 same year, order not determinable · 5 only as an assembly brick
1409.0473 (2014-09)
Neural Machine Translation by Jointly Learning to Align and Translate
1706.03762 (2017-06)
Attention Is All You Need
1904.01681 (2019-04)
Augmented Neural ODEs
and 55 more not named here
Embedding 29 other papers
27 dated earlier, 1 later, 1 same year, order not determinable
1904.01681 (2019-04)
Augmented Neural ODEs
1906.12183 (2019-06)
Neural ODEs as the Deep Limit of ResNets with constant weights
2002.03184 (2020-02)
Time-aware Large Kernel Convolutions
and 26 more not named here
apply_half 6 other papers
6 dated earlier
2202.03052 (2022-02)
OFA: Unifying Architectures, Tasks, and Modalities Through a Simple Sequence-to-Sequence Learning Framework
2208.02532 (2022-08)
Prompt Tuning for Generative Multimodal Pretrained Models
2212.04231 (2022-12)
Harnessing the Power of Multi-Task Pretraining for Ground-Truth Level Natural Language Explanations
and 3 more not named here
downsample_basic_block 5 other papers
4 dated earlier, 1 later
2312.02512 (2023-12)
AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
2401.03468 (2024-01)
Multichannel AV-wav2vec2: A Framework for Learning Multichannel Multi-Modal Speech Representation
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
and 2 more not named here
downsample_basic_block_v2 5 other papers
4 dated earlier, 1 later
2312.02512 (2023-12)
AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
2401.03468 (2024-01)
Multichannel AV-wav2vec2: A Framework for Learning Multichannel Multi-Modal Speech Representation
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
and 2 more not named here
load_label 4 other papers
3 dated earlier, 1 later
2312.02512 (2023-12)
AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
2401.03468 (2024-01)
Multichannel AV-wav2vec2: A Framework for Learning Multichannel Multi-Modal Speech Representation
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
and 1 more not named here
load_label_offset 4 other papers
3 dated earlier, 1 later
2312.02512 (2023-12)
AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
2401.03468 (2024-01)
Multichannel AV-wav2vec2: A Framework for Learning Multichannel Multi-Modal Speech Representation
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
and 1 more not named here
get_symbols_to_strip_from_output 3 other papers
2 dated earlier, 1 later
1909.08723 (2019-09)
Espresso: A Fast End-to-end Neural Speech Recognition Toolkit
2201.01763 (2022-01)
Robust Self-Supervised Audio-Visual Speech Recognition
2504.18539 (2025-04)
Multi-Task Corrupted Prediction for Learning Robust Audio-Visual Speech Representation
load_audio_visual 3 other papers
2 dated earlier, 1 later
2312.02512 (2023-12)
AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
2602.21646 (2026-02)
Scalable Multilingual Multimodal Machine Translation with Speech-Text Fusion
compute_mask_indices 1 other paper
1 dated earlier
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
find_runs 1 other paper
1 dated earlier
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing
load_video 1 other paper
1 dated earlier
2402.15151 (2024-02)
Where Visual Speech Meets Language: VSP-LLM Framework for Efficient and Context-Aware Visual Speech Processing

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 (AVSR) aims to transcribe human speech using both audio and video modalities. In practical environments with noise-corrupted audio, the role of video information becomes crucial. However, prior works have primarily focused on enhancing audio features in AVSR, overlooking the importance of video features. In this study, we strengthen the video features by learning three temporal dynamics in video data: context order, playback direction, and the speed of video frames. Cross-modal attention modules are introduced to enrich video features with audio information so that speech variability can be taken into account when training on the video temporal dynamics. Based on our approach, we achieve the state-of-the-art performance on the LRS2 and LRS3 AVSR benchmarks for the noise-dominant settings. Our approach excels in scenarios especially for babble and speech noise, indicating the ability to distinguish the speech signal that should be recognized from lip movements in the video modality. We support the validity of our methodology by offering the ablation experiments for the temporal dynamics losses and the cross-modal attention architecture design.

For agents

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

get_harvested_code_for_paper("2407.03563")
get_code_for_paper("2407.03563")
have("2407.03563")

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

curl https://syntology.ai/api/ran/2407.03563.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.03563.svg)](https://syntology.ai/paper/2407.03563)

Connect an agent — have() is free.