SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1611.01599 · 2016

LipNet: End-to-End Sentence-level Lipreading

arXiv · PDF · Open in the Atlas

Code that ran

10 of 15 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 9 have no contract verdict; 1 failure of our instrument; 5 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/1611.01599.json · badge: /api/ran/1611.01599.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
rizkiarm/LipNet canonical (classifier label)
classifier basis: census
3 ran · 2 no recorded run
pjenpoomjai/LipNet community (classifier label)
classifier basis not recorded on the link
6 ran · 0 no recorded run
sailordiary/LipNet-PyTorch community (classifier label)
classifier basis not recorded on the link
1 ran · 3 no recorded run
FunctionStatusWhere it lives
get_file_extension Ran
ran on a synthesized input
pjenpoomjai/LipNet/common/files.py
code served (permissive licence) · get_code("bc11ac6762ddee1c")
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 os
import tempfile

# Create a temporary file for testing
with tempfile.NamedTemporaryFile(mode='w', suffix='.txt', delete=False) as f:
    temp_file_path = f.name
    f.write("test content")

ARGS = {
    "path": temp_file_path
}
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 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
is_dir Ran
ran on a synthesized input
pjenpoomjai/LipNet/common/files.py
code served (permissive licence) · get_code("0ab02e700216530c")
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 os
import tempfile
import shutil

# Create a temporary directory for testing
temp_dir = tempfile.mkdtemp()

# Create a subdirectory to test the is_dir function
test_dir = os.path.join(temp_dir, "test_directory")
os.makedirs(test_dir, exist_ok=True)

# Define arguments for the is_dir function
ARGS = {
    "path": test_dir
}

# Cleanup will happen when the process exits, but we could also do:
# shutil.rmtree(temp_dir)
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 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
is_file Ran
ran on a synthesized input
pjenpoomjai/LipNet/common/files.py
code served (permissive licence) · get_code("74ffcb74cead6878")
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 os
import tempfile

# Create a temporary file for testing
with tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt') as f:
    temp_file_path = f.name
    f.write("test content")

ARGS = {
    "path": temp_file_path
}
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 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
labels_to_text Ran
ran on a synthesized input
rizkiarm/LipNet/lipnet/lipreading/helpers.py
code served (permissive licence) · get_code("ba5f0f129eba0ac3")
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

torch.manual_seed(0)

# The function takes labels (iterable of integers) and converts them to text
# where 0-25 map to 'a'-'z' and 26 maps to ' '
# We need to construct a valid labels argument

labels = torch.tensor([0, 1, 25, 26, 5, 14, 7], dtype=torch.long)

ARGS = {
    'labels': labels
}
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
round Ran
ran on a synthesized input
sailordiary/LipNet-PyTorch/augmentation.py
code served (permissive licence) · get_code("6460635662358375")
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 math

def round(x):
    return math.floor(x + 0.5)

# The function takes a single parameter 'x' and applies math.floor(x + 0.5)
# This works with any numeric type (int, float)

ARGS = {
    "x": 3.7
}
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 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
text_to_labels Ran
ran on a synthesized input
rizkiarm/LipNet/lipnet/lipreading/helpers.py
code served (permissive licence) · get_code("071ed9f14cb9d384")
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 list, 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.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "text": "hello world"
}
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 list, ndim 1, dtype family int — 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
threadsafe_generator Ran
ran; our contract draft was wrong, not the code
rizkiarm/LipNet/lipnet/helpers/threadsafe.py
code served (permissive licence) · get_code("6656877b73807105")
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 threading
import torch

torch.manual_seed(0)

class threadsafe_iter:
    """Takes an iterator/generator and makes it thread-safe by
    serializing call to the `next` method of given iterator/generator.
    """

    def __init__(self, it):
        self.it = it
        self.lock = threading.Lock()

    def __iter__(self):
        return self

    def next(self):
        with self.lock:
            return self.it.next()

def threadsafe_generator(f):
    """A decorator that takes a generator function and makes it thread-safe.
    """

    def g(*a, **kw):
        return threadsafe_iter(f(*a, **kw))
    return g


# Create a simple generator function to pass as the decorator argument
def sample_generator(n):
    """A simple generator that yields n values."""
    for i in range(n):
        yield torch.randn(2, 4)

# The threadsafe_generator function takes a generator function f as its parameter
# It returns a decorator that wraps that generator function
ARGS = {
    "f": sample_generator
}
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
tokenize Ran
ran on a synthesized input
pjenpoomjai/LipNet/core/decoding/spell.py
code served (permissive licence) · get_code("9108c44188636656")
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 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 re

def tokenize(text: str) -> list:
    return re.findall('\\w+|[^\\w\\s]', text, re.UNICODE)

ARGS = {
    "text": "Hello, world! This is a test 123."
}
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 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
untokenize Ran
ran on a synthesized input
pjenpoomjai/LipNet/core/decoding/spell.py
code served (permissive licence) · get_code("8b832c2b8bf29a51")
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 re

def untokenize(words: list) -> str:
    """
	Untokenizing a text undoes the tokenizing operation, restoring
	punctuation and spaces to the places that people expect them to be.
	Ideally, `untokenize(tokenize(text))` should be identical to `text`,
	except for line breaks.
	"""
    text = ' '.join(words)
    step1 = text.replace('`` ', '"').replace(" ''", '"').replace('. . .', '...')
    step2 = step1.replace(' ( ', ' (').replace(' ) ', ') ')
    step3 = re.sub(' ([.,:;?!%]+)([ \\\'"`])', '\\1\\2', step2)
    step4 = re.sub(' ([.,:;?!%]+)$', '\\1', step3)
    step5 = step4.replace(" '", "'").replace(" n't", "n't").replace('can not', 'cannot')
    step6 = step5.replace(' ` ', " '")
    return step6.strip()

ARGS = {
    "words": ["Hello", "world", "!", "This", "is", "a", "test", "."]
}
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 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
video_to_frames Ran
ran on a synthesized input
pjenpoomjai/LipNet/core/video.py
code served (permissive licence) · get_code("face0a3bb9c1c095")
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 cv2
import numpy as np
import tempfile
import os

# Create a temporary video file for testing
def create_test_video(filename, num_frames=10, width=640, height=480):
    """Create a simple test video file"""
    fourcc = cv2.VideoWriter_fourcc(*'mp4v')
    out = cv2.VideoWriter(filename, fourcc, 30.0, (width, height))
    
    for i in range(num_frames):
        # Create a simple frame with changing colors
        frame = np.zeros((height, width, 3), dtype=np.uint8)
        frame[:, :, 0] = (i * 25) % 256  # Blue channel varies
        frame[:, :, 1] = (i * 50) % 256  # Green channel varies
        frame[:, :, 2] = (i * 75) % 256  # Red channel varies
        out.write(frame)
    
    out.release()

# Create temporary video file
temp_dir = tempfile.gettempdir()
video_path = os.path.join(temp_dir, "test_video.mp4")
create_test_video(video_path, num_frames=10)

ARGS = {
    "video_filename": 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 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
ctc_collate Not yet run
unverified (harvested, no recorded run)
sailordiary/LipNet-PyTorch/dataloader.py
code served (permissive licence) · get_code("88b8acae6777424e")
How it ran

no invocation recorded.

ctc_lambda_func Not yet run
unverified (harvested, no recorded run)
rizkiarm/LipNet/lipnet/core/loss.py
code served (permissive licence) · get_code("8181d94a6381a30a")
How it ran

no invocation recorded.

decode Not yet run
unverified (harvested, no recorded run)
rizkiarm/LipNet/lipnet/core/decoders.py
code served (permissive licence) · get_code("07d26c38ea615861")
How it ran

no invocation recorded.

read_data Not yet run
unverified (harvested, no recorded run)
sailordiary/LipNet-PyTorch/augmentation.py
code served (permissive licence) · get_code("f3ec786915134721")
How it ran

no invocation recorded.

round Not yet run
unverified (harvested, no recorded run)
sailordiary/LipNet-PyTorch/dataloader.py
code served (permissive licence) · get_code("80b5790c95657203")
How it ran

no invocation recorded.

Also present in other papers' code

Of 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 13 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
threadsafe_generator 1 other paper
1 later
1612.01051 (2016-12)
SqueezeDet: Unified, Small, Low Power Fully Convolutional Neural Networks for Real-Time Object Detection for Autonomous Driving

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

Lipreading is the task of decoding text from the movement of a speaker's mouth. Traditional approaches separated the problem into two stages: designing or learning visual features, and prediction. More recent deep lipreading approaches are end-to-end trainable (Wand et al., 2016; Chung & Zisserman, 2016a). However, existing work on models trained end-to-end perform only word classification, rather than sentence-level sequence prediction. Studies have shown that human lipreading performance increases for longer words (Easton & Basala, 1982), indicating the importance of features capturing temporal context in an ambiguous communication channel. Motivated by this observation, we present LipNet, a model that maps a variable-length sequence of video frames to text, making use of spatiotemporal convolutions, a recurrent network, and the connectionist temporal classification loss, trained entirely end-to-end. To the best of our knowledge, LipNet is the first end-to-end sentence-level lipreading model that simultaneously learns spatiotemporal visual features and a sequence model. On the GRID corpus, LipNet achieves 95.2% accuracy in sentence-level, overlapped speaker split task, outperforming experienced human lipreaders and the previous 86.4% word-level state-of-the-art accuracy (Gergen et al., 2016).

For agents

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

get_harvested_code_for_paper("1611.01599")
get_code_for_paper("1611.01599")
have("1611.01599")

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

curl https://syntology.ai/api/ran/1611.01599.json

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

[![Syntology run record](https://syntology.ai/api/ran/1611.01599.svg)](https://syntology.ai/paper/1611.01599)

Connect an agent — have() is free.