SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2105.02446 · 2021

DiffSinger: Singing Voice Synthesis via Shallow Diffusion Mechanism

arXiv · PDF · Open in the Atlas

Code that ran

7 of 7 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; 1 honoured the contract we drafted, 1 violated it and 3 have no contract verdict; 2 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2105.02446.json · badge: /api/ran/2105.02446.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
copy not recorded community
not in the paper's declared repositories
4 ran · 0 no recorded run
openvpi/DiffSinger community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
FunctionStatusWhere it lives
amp_to_db Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from AaronZ345/GTSinger
pointer only; the stored copy get_code serves, from AaronZ345/GTSinger, is pointer only (licence: NOASSERTION) · get_code("3c41caef21862a65")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

get_backbone_args Ran
ran on a synthesized input
openvpi/DiffSinger/modules/compat.py
code served (permissive licence) · get_code("1812d15f57a983c9")
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 torch.nn as nn

torch.manual_seed(0)

# Construct arguments for get_backbone_args function

# Test case 1: config with backbone_args already defined
config_with_args = {
    'backbone_args': {'num_layers': 10, 'num_channels': 64},
    'residual_layers': 5,
    'residual_channels': 32,
    'dilation_cycle_length': 2
}
backbone_type_1 = 'wavenet'

# Test case 2: config without backbone_args, wavenet backbone type
config_wavenet = {
    'residual_layers': 8,
    'residual_channels': 128,
    'dilation_cycle_length': 3
}
backbone_type_2 = 'wavenet'

# Test case 3: config without backbone_args, other backbone type
config_other = {
    'residual_layers': 6,
    'residual_channels': 64,
    'dilation_cycle_length': 2
}
backbone_type_3 = 'resnet'

# Use the first test case for ARGS
ARGS = {
    'config': config_with_args,
    'backbone_type': backbone_type_1
}
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
get_backbone_type Ran
ran on a synthesized input
openvpi/DiffSinger/modules/compat.py
code served (permissive licence) · get_code("3252454123d7ae1d")
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

def get_backbone_type(root_config: dict, nested_config: dict=None):
    if nested_config is None:
        nested_config = root_config
    return nested_config.get('backbone_type', root_config.get('backbone_type', root_config.get('diff_decoder_type', 'wavenet')))

# Construct valid arguments
root_config = {
    'backbone_type': 'resnet',
    'diff_decoder_type': 'wavenet',
    'other_key': 'other_value'
}

nested_config = {
    'backbone_type': 'vit',
    'some_param': 42
}

ARGS = {
    'root_config': root_config,
    'nested_config': nested_config
}
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 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
get_hop_size Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from AaronZ345/GTSinger
pointer only; the stored copy get_code serves, from AaronZ345/GTSinger, is pointer only (licence: NOASSERTION) · get_code("12fb4d2bf6cdf9c4")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

librosa_pad_lr Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
this paper's copy was not recorded; identical code first harvested from AaronZ345/GTSinger
pointer only; the stored copy get_code serves, from AaronZ345/GTSinger, is pointer only (licence: NOASSERTION) · get_code("9711d1038efa8793")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

read_lexicon Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from KevinMIN95/StyleSpeech
pointer only · get_code("2b5c06c1ba593c3c")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

require_same_keys Ran
ran on a synthesized input
openvpi/DiffSinger/basics/base_augmentation.py
code served (permissive licence) · get_code("b8c7b14829fa9e1a")
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 torch
import torch.nn as nn

torch.manual_seed(0)

def require_same_keys(func):
    def run(*args, **kwargs):
        item: dict = args[1]
        res: dict = func(*args, **kwargs)
        assert set(item.keys()) == set(res.keys()), f'Item keys mismatch after augmentation.\nBefore: {sorted(item.keys())}\nAfter: {sorted(res.keys())}'
        return res
    return run

# Create a simple function that will be decorated
@require_same_keys
def augment_function(self, item):
    # Function that preserves keys
    result = {}
    for key, value in item.items():
        if isinstance(value, torch.Tensor):
            result[key] = value * 2
        else:
            result[key] = value
    return result

# Create a simple class to hold the augment function
class Augmenter:
    @require_same_keys
    def augment(self, item):
        result = {}
        for key, value in item.items():
            if isinstance(value, torch.Tensor):
                result[key] = value * 2
            else:
                result[key] = value
        return result

# Create test data
augmenter = Augmenter()

test_item = {
    'image': torch.randn(2, 4),
    'label': torch.tensor([0, 1]),
    'name': 'test_sample'
}

ARGS = {
    'func': augmenter.augment,
}
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

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 4 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 4 later) and 3 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
read_lexicon 11 other papers
4 dated earlier, 1 the same month, 6 later
1905.09263 (2019-05)
FastSpeech: Fast, Robust and Controllable Text to Speech
2006.04558 (2020-06)
FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
2006.06873 (2020-06)
FastPitch: Parallel Text-to-speech with Pitch Prediction
and 8 more not named here
amp_to_db 1 other paper
1 later
2409.13832 (2024-09)
GTSinger: A Global Multi-Technique Singing Corpus with Realistic Music Scores for All Singing Tasks
get_hop_size 1 other paper
1 later
2409.13832 (2024-09)
GTSinger: A Global Multi-Technique Singing Corpus with Realistic Music Scores for All Singing Tasks
librosa_pad_lr 1 other paper
1 later
2409.13832 (2024-09)
GTSinger: A Global Multi-Technique Singing Corpus with Realistic Music Scores for All Singing Tasks

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Singing voice synthesis (SVS) systems are built to synthesize high-quality and expressive singing voice, in which the acoustic model generates the acoustic features (e.g., mel-spectrogram) given a music score. Previous singing acoustic models adopt a simple loss (e.g., L1 and L2) or generative adversarial network (GAN) to reconstruct the acoustic features, while they suffer from over-smoothing and unstable training issues respectively, which hinder the naturalness of synthesized singing. In this work, we propose DiffSinger, an acoustic model for SVS based on the diffusion probabilistic model. DiffSinger is a parameterized Markov chain that iteratively converts the noise into mel-spectrogram conditioned on the music score. By implicitly optimizing variational bound, DiffSinger can be stably trained and generate realistic outputs. To further improve the voice quality and speed up inference, we introduce a shallow diffusion mechanism to make better use of the prior knowledge learned by the simple loss. Specifically, DiffSinger starts generation at a shallow step smaller than the total number of diffusion steps, according to the intersection of the diffusion trajectories of the ground-truth mel-spectrogram and the one predicted by a simple mel-spectrogram decoder. Besides, we propose boundary prediction methods to locate the intersection and determine the shallow step adaptively. The evaluations conducted on a Chinese singing dataset demonstrate that DiffSinger outperforms state-of-the-art SVS work. Extensional experiments also prove the generalization of our methods on text-to-speech task (DiffSpeech). Audio samples: https://diffsinger.github.io. Codes: https://github.com/MoonInTheRiver/DiffSinger. The old title of this work: "Diffsinger: Diffusion acoustic model for singing voice synthesis".

For agents

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

get_harvested_code_for_paper("2105.02446")
get_code_for_paper("2105.02446")
have("2105.02446")

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

curl https://syntology.ai/api/ran/2105.02446.json

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

[![Syntology run record](https://syntology.ai/api/ran/2105.02446.svg)](https://syntology.ai/paper/2105.02446)

Connect an agent — have() is free.