SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2307.02471 · 2023

Deep Speech Synthesis from MRI-Based Articulatory Representations

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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; 1 failure 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/2307.02471.json · badge: /api/ran/2307.02471.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
articulatory/articulatory canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
ar_loop Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
articulatory/articulatory/articulatory/bin/decode.py
code served (permissive licence) · get_code("bcd6d3cede5039f0")
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 1, dtype family float, shape (24,) 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 torch.nn as nn

torch.manual_seed(0)

# Build a minimal model that satisfies the function's usage
class MinimalGenerator(nn.Module):
    def __init__(self, out_channels=2, in_channels=1):
        super().__init__()
        self.out_channels = out_channels
        self.in_channels = in_channels
        self.linear = nn.Linear(in_channels, out_channels)
    
    def forward(self, cin, ar=None):
        # cin can be a tensor or a list (when modality is not None)
        if isinstance(cin, list):
            # Use the first non-None element
            x = cin[0]
        else:
            x = cin
        
        # x shape: (batch, in_channels, seq_len)
        batch, channels, seq_len = x.shape
        x = x.permute(0, 2, 1)  # (batch, seq_len, in_channels)
        x = self.linear(x)  # (batch, seq_len, out_channels)
        x = x.permute(0, 2, 1)  # (batch, out_channels, seq_len)
        
        # Expand to match expected output length
        if ar is not None:
            ar_len = ar.shape[2]
            # Output should be (batch, out_channels, ar_len + seq_len) or similar
            output = torch.cat([ar, x], dim=2)
        else:
            output = x
        
        return output

# Create config dict
config = {
    'batch_max_steps': 8,
    'hop_size': 2,
    'sampling_rate': 16000,
    'dataset_mode': 'a2a',  # not 'w2a'
    'generator_params': {
        'ar_input': 4,
        'out_channels': 2,
        'in_list': [0],
        'extra_art': 2,
    },
    'generator2_params': {
        'ar_input': 4,
        'out_channels': 2,
        'in_list': [0],
        'extra_art': 2,
    },
    'hop_sizes': [2],
    'sampling_rates': [16000],
}

# Create model
model = MinimalGenerator(out_channels=2, in_channels=1)

# Create input tensor: (art_len, num_feats)
# art_len should be divisible by in_chunk_len for clean chunking
# in_chunk_len = audio_chunk_len / hop_size = 8 / 2 = 4
x = torch.randn(12, 1, dtype=torch.float32)  # 12 samples, 1 feature

ARGS = {
    'model': model,
    'x': x,
    'config': config,
    'do_wsola': False,
    'modality': None,
    'generator2': False,
}
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 1, dtype family float, shape (24,) — shape (24,) 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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 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.

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

In this paper, we study articulatory synthesis, a speech synthesis method using human vocal tract information that offers a way to develop efficient, generalizable and interpretable synthesizers. While recent advances have enabled intelligible articulatory synthesis using electromagnetic articulography (EMA), these methods lack critical articulatory information like excitation and nasality, limiting generalization capabilities. To bridge this gap, we propose an alternative MRI-based feature set that covers a much more extensive articulatory space than EMA. We also introduce normalization and denoising procedures to enhance the generalizability of deep learning methods trained on MRI data. Moreover, we propose an MRI-to-speech model that improves both computational efficiency and speech fidelity. Finally, through a series of ablations, we show that the proposed MRI representation is more comprehensive than EMA and identify the most suitable MRI feature subset for articulatory synthesis.

For agents

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

get_harvested_code_for_paper("2307.02471")
get_code_for_paper("2307.02471")
have("2307.02471")

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

curl https://syntology.ai/api/ran/2307.02471.json

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

[![Syntology run record](https://syntology.ai/api/ran/2307.02471.svg)](https://syntology.ai/paper/2307.02471)

Connect an agent — have() is free.