SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.18503 · 2024

SoundCTM: Unifying Score-based and Consistency Models for Full-band Text-to-Sound Generation

arXiv · PDF · Open in the Atlas

Code that ran

2 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; 2 failures of our instrument; 3 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2405.18503.json · badge: /api/ran/2405.18503.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
sony/soundctm canonical (classifier label)
classifier basis: census
2 ran · 3 no recorded run
FunctionStatusWhere it lives
append_dims Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sony/soundctm/tango_edm/models_edm.py
pointer only (licence: MIT); the stored copy get_code serves, from layer6ai-labs/direct-cms, is pointer only (licence: NONE) · get_code("30befb7e4327e615")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from layer6ai-labs/direct-cms, 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_state_dict Ran
ran; our contract draft was wrong, not the code
sony/soundctm/clap/clap_module/factory.py
code served (permissive licence) · get_code("6ce126db645cc0a7")
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
import tempfile
import os

torch.manual_seed(0)

# Create a temporary checkpoint file with a valid state_dict
temp_dir = tempfile.mkdtemp()
checkpoint_path = os.path.join(temp_dir, "checkpoint.pt")

# Create a sample state_dict with the structure the function expects
state_dict = {
    'module.text_branch.embeddings.position_ids': torch.tensor([0, 1, 2]),
    'text_branch.embeddings.position_ids': torch.tensor([0, 1, 2]),
    'layer1.weight': torch.randn(4, 3),
    'layer1.bias': torch.randn(4),
}

# Save checkpoint in two formats to test both branches
checkpoint_with_wrapper = {
    'state_dict': state_dict,
    'epoch': 1,
}

torch.save(checkpoint_with_wrapper, checkpoint_path)

ARGS = {
    'checkpoint_path': checkpoint_path,
    'map_location': 'cpu',
    'skip_params': True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
packaging, torch, transformers — 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
create_model Not yet run
unverified (harvested, no recorded run)
sony/soundctm/clap/clap_module/factory.py
code served (permissive licence) · get_code("3ab23e5d9f9352a8")
How it ran

no invocation recorded.

create_model_and_transforms Not yet run
unverified (harvested, no recorded run)
sony/soundctm/clap/clap_module/factory.py
code served (permissive licence) · get_code("bfdd273dab42b860")
How it ran

no invocation recorded.

edm_sampler Not yet run
unverified (harvested, no recorded run)
sony/soundctm/tango_edm/models_edm.py
code served (permissive licence) · get_code("8b16b0674af254db")
How it ran

no invocation recorded.

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 (4 with at least one paper dated earlier, 1 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
append_dims 27 other papers
13 dated earlier, 3 the same month, 9 later, 2 same year, order not determinable · 2 only as an assembly brick
2204.06125 (2022-04)
Hierarchical Text-Conditional Image Generation with CLIP Latents
2304.09787 (2023-04)
NeuralField-LDM: Scene Generation with Hierarchical Latent Diffusion Models
2307.12493 (2023-07)
TF-ICON: Diffusion-Based Training-Free Cross-Domain Image Composition
and 24 more not named here
edm_sampler 11 other papers
7 dated earlier, 4 later
2206.00364 (2022-06)
Elucidating the Design Space of Diffusion-Based Generative Models
2301.11445 (2023-01)
3DShape2VecSet: A 3D Shape Representation for Neural Fields and Generative Diffusion Models
2303.10137 (2023-03)
A Recipe for Watermarking Diffusion Models
and 8 more not named here
create_model_and_transforms 4 other papers
2 dated earlier, 2 later
2308.01546 (2023-08)
MusicLDM: Enhancing Novelty in Text-to-Music Generation Using Beat-Synchronous Mixup Strategies
2404.01740 (2024-04)
Weakly-supervised Audio Separation via Bi-modal Semantic Similarity
2409.12346 (2024-09)
Simultaneous Music Separation and Generation Using Multi-Track Latent Diffusion Models
and 1 more not named here
create_model 2 other papers
1 dated earlier, 1 later
2404.01740 (2024-04)
Weakly-supervised Audio Separation via Bi-modal Semantic Similarity
2605.30107 (2026-05)
Dial HEALTHDIAL for Advice: A Multilingual and Multi-Parallel Spoken Dialogue Dataset for Knowledge-Grounded Information Seeking

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

Sound content creation, essential for multimedia works such as video games and films, often involves extensive trial-and-error, enabling creators to semantically reflect their artistic ideas and inspirations, which evolve throughout the creation process, into the sound. Recent high-quality diffusion-based Text-to-Sound (T2S) generative models provide valuable tools for creators. However, these models often suffer from slow inference speeds, imposing an undesirable burden that hinders the trial-and-error process. While existing T2S distillation models address this limitation through 1-step generation, the sample quality of $1$-step generation remains insufficient for production use. Additionally, while multi-step sampling in those distillation models improves sample quality itself, the semantic content changes due to their lack of deterministic sampling capabilities. To address these issues, we introduce Sound Consistency Trajectory Models (SoundCTM), which allow flexible transitions between high-quality $1$-step sound generation and superior sound quality through multi-step deterministic sampling. This allows creators to efficiently conduct trial-and-error with 1-step generation to semantically align samples with their intention, and subsequently refine sample quality with preserving semantic content through deterministic multi-step sampling. To develop SoundCTM, we reframe the CTM training framework, originally proposed in computer vision, and introduce a novel feature distance using the teacher network for a distillation loss. For production-level generation, we scale up our model to 1B trainable parameters, making SoundCTM-DiT-1B the first large-scale distillation model in the sound community to achieve both promising high-quality 1-step and multi-step full-band (44.1kHz) generation.

For agents

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

get_harvested_code_for_paper("2405.18503")
get_code_for_paper("2405.18503")
have("2405.18503")

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

curl https://syntology.ai/api/ran/2405.18503.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.18503.svg)](https://syntology.ai/paper/2405.18503)

Connect an agent — have() is free.