SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.01748 · 2024

NeuSpeech: Decode Neural signal as Speech

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 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 0 have no contract verdict; 5 failures of our instrument; 0 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/2403.01748.json · badge: /api/ran/2403.01748.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
neuspeech/neuspeech1 canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_accuracy Ran
ran; our contract draft was wrong, not the code
neuspeech/neuspeech1/utils/model_utils.py
code served (permissive licence) · get_code("262947201dc066e9")
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)

class PredictionResult:
    def __init__(self, predictions, label_ids):
        self.predictions = predictions
        self.label_ids = label_ids

# Create small tensors
# predictions[0] shape: (batch_size=2, seq_length=3, num_classes=4)
predictions_logits = torch.randn(2, 3, 4)

# label_ids shape: (batch_size=2, seq_length=3)
label_ids = torch.randint(0, 4, (2, 3))

# Create prediction object with required attributes
pred = PredictionResult(
    predictions=[predictions_logits.numpy()],
    label_ids=label_ids.numpy()
)

ARGS = {
    'pred': pred
}
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
compute_metrics Ran
ran; our contract draft was wrong, not the code
neuspeech/neuspeech1/metrics/bleu.py
code served (permissive licence) · get_code("b1936507e543343e")
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)

# Looking at the function, it expects:
# - preds: a list of predicted strings
# - labels: a list of reference strings (will be wrapped in lists)

# BLEUScore expects preds to be a list of strings and labels to be a list of lists of strings
# The function will wrap each label in a list, so we just need string labels

preds = [
    "the quick brown fox jumps over the lazy dog",
    "a cat sat on the mat"
]

labels = [
    "the quick brown fox jumps over the lazy dog",
    "the cat is sitting on the mat"
]

ARGS = {
    "preds": preds,
    "labels": labels
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchmetrics — 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
match_modules Ran
ran; our contract draft was wrong, not the code
neuspeech/neuspeech1/utils/load_model.py
code served (permissive licence) · get_code("bf84bf80d676a07a")
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 torch
import torch.nn as nn
import re

torch.manual_seed(0)

# Build a small nn.Module to use as a stand-in
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        return self.linear2(self.linear1(x))

model = SimpleModel()

# Get named_modules - this is what the function iterates over
# named_modules returns an iterator of (name, module) tuples
named_modules_list = list(model.named_modules())

# Create prefix_list with some patterns that will match module names
prefix_list = ['', 'linear']

# Create suffix_list with some patterns
suffix_list = ['', '1', '2']

# Create mid_fix_list with some patterns (optional, defaults to [''])
mid_fix_list = ['']

ARGS = {
    'named_modules': named_modules_list,
    'prefix_list': prefix_list,
    'suffix_list': suffix_list,
    'mid_fix_list': mid_fix_list
}
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 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
match_modules_string Ran
ran; our contract draft was wrong, not the code
neuspeech/neuspeech1/utils/load_model.py
code served (permissive licence) · get_code("1c6ebf4c6f313f2c")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Build a small model with named modules
class SmallModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder_layer1 = nn.Linear(4, 3)
        self.encoder_layer2_weight = nn.Linear(3, 2)
        self.decoder_layer1 = nn.Linear(2, 3)
        self.decoder_attention = nn.Linear(3, 4)
        self.head_output = nn.Linear(4, 2)
        
model = SmallModel()

# Get named_modules from the model
# This returns an iterator of (name, module) tuples
named_modules = list(model.named_modules())

# Define test arguments
start_prefixes = ["encoder", "decoder"]
end_suffixes = ["weight", "output"]
mid_prefixes = ["layer"]

ARGS = {
    "named_modules": named_modules,
    "start_prefixes": start_prefixes,
    "end_suffixes": end_suffixes,
    "mid_prefixes": mid_prefixes
}
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 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
projection_module Ran
ran; our contract draft was wrong, not the code
neuspeech/neuspeech1/utils/model_utils.py
code served (permissive licence) · get_code("db94765af5a0e4dd")
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
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'config_name': 'base',
    'd_model': 64,
    'meg_ch': 32
}
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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 4 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 4 later) and 1 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
compute_accuracy 1 other paper
1 later
2406.01512 (2024-06)
MAD: Multi-Alignment MEG-to-Text Decoding
match_modules 1 other paper
1 later
2406.01512 (2024-06)
MAD: Multi-Alignment MEG-to-Text Decoding
match_modules_string 1 other paper
1 later
2406.01512 (2024-06)
MAD: Multi-Alignment MEG-to-Text Decoding
projection_module 1 other paper
1 later
2406.01512 (2024-06)
MAD: Multi-Alignment MEG-to-Text Decoding

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

Decoding language from brain dynamics is an important open direction in the realm of brain-computer interface (BCI), especially considering the rapid growth of large language models. Compared to invasive-based signals which require electrode implantation surgery, non-invasive neural signals (e.g. EEG, MEG) have attracted increasing attention considering their safety and generality. However, the exploration is not adequate in three aspects: 1) previous methods mainly focus on EEG but none of the previous works address this problem on MEG with better signal quality; 2) prior works have predominantly used $``teacher-forcing"$ during generative decoding, which is impractical; 3) prior works are mostly $``BART-based"$ not fully auto-regressive, which performs better in other sequence tasks. In this paper, we explore the brain-to-text translation of MEG signals in a speech-decoding formation. Here we are the first to investigate a cross-attention-based ``whisper" model for generating text directly from MEG signals without teacher forcing. Our model achieves impressive BLEU-1 scores of 60.30 and 52.89 without pretraining $\&$ teacher-forcing on two major datasets ($\textit{GWilliams}$ and $\textit{Schoffelen}$). This paper conducts a comprehensive review to understand how speech decoding formation performs on the neural decoding tasks, including pretraining initialization, training $\&$ evaluation set splitting, augmentation, and scaling law. Code is available at https://github.com/NeuSpeech/NeuSpeech1$.

For agents

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

get_harvested_code_for_paper("2403.01748")
get_code_for_paper("2403.01748")
have("2403.01748")

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

curl https://syntology.ai/api/ran/2403.01748.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.01748.svg)](https://syntology.ai/paper/2403.01748)

Connect an agent — have() is free.