SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.13975 · 2020

Dual-Path Transformer Network: Direct Context-Aware Modeling for End-to-End Monaural Speech Separation

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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; 1 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/2007.13975.json · badge: /api/ran/2007.13975.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
saurjya/asteroid community (classifier label)
classifier basis not recorded on the link
5 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
FunctionStatusWhere it lives
batch_matrix_norm Ran
ran on a synthesized input
saurjya/asteroid/asteroid/losses/cluster.py
code served (permissive licence) · get_code("b191df325b6d1eb4")
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 Tensor, ndim 1, dtype family float, shape (2,) 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

torch.manual_seed(0)

# Create a small batch of matrices
# Shape: [batch=2, height=4, width=3]
matrix = torch.randn(2, 4, 3)

# norm_order parameter
norm_order = 2

ARGS = {
    "matrix": matrix,
    "norm_order": norm_order,
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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
cal_SISNR Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from kaituoxu/Conv-TasNet
pointer only · get_code("faac9c0b7e1a08c5")
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.

cal_SISNRi Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from kaituoxu/Conv-TasNet
pointer only · get_code("b19791aa06bb7b9b")
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.

deep_clustering_loss Ran
ran; our contract draft was wrong, not the code
saurjya/asteroid/asteroid/losses/cluster.py
code served (permissive licence) · get_code("9ad382778f11ca4c")
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 Tensor, ndim 1, dtype family float, shape (2,) 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

torch.manual_seed(0)

# Parameters for deep_clustering_loss
batch = 2
freq = 4
frames = 3
embedding_dim = 5
spk_cnt = 2

# embedding: shape (batch, frequency * frame, embedding_dim)
embedding = torch.randn(batch, freq * frames, embedding_dim)

# tgt_index: shape (batch, frequency, frame) with values in [0, spk_cnt)
tgt_index = torch.randint(0, spk_cnt, (batch, freq, frames))

# binary_mask: shape (batch, frequency * frame, 1) - optional, can be None
binary_mask = torch.ones(batch, freq * frames, 1)

ARGS = {
    "embedding": embedding,
    "tgt_index": tgt_index,
    "binary_mask": binary_mask,
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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_torch_complex Ran
ran on a synthesized input
saurjya/asteroid/asteroid/complex_nn.py
code served (permissive licence) · get_code("0445a31bc868218d")
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 torch

torch.manual_seed(0)

# The function calls x.is_complex() on the argument x
# So x needs to be a torch tensor with the is_complex() method
# Create a complex tensor to test the function
complex_tensor = torch.complex(torch.randn(2, 4), torch.randn(2, 4))

ARGS = {
    "x": complex_tensor
}
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 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
torch_complex_from_magphase Ran
ran on a synthesized input
saurjya/asteroid/asteroid/complex_nn.py
code served (permissive licence) · get_code("d695a0f6cb2e24e3")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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

torch.manual_seed(0)

# Create small tensors for magnitude and phase
# mag: magnitude values (should be non-negative)
# phase: phase values in radians
mag = torch.rand(2, 4) * 2.0  # Small random magnitudes [0, 2)
phase = torch.rand(2, 4) * 2 * torch.pi  # Random phases [0, 2π)

ARGS = {
    "mag": mag,
    "phase": phase,
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
torch_complex_from_reim Ran
ran; our contract draft was wrong, not the code
saurjya/asteroid/asteroid/complex_nn.py
code served (permissive licence) · get_code("63cec12b7bc5275c")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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)

# Create small real and imaginary tensors
re = torch.randn(2, 4)
im = torch.randn(2, 4)

ARGS = {
    "re": re,
    "im": im,
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
make_unet_encoder_decoder_args Not yet run
unverified (harvested, no recorded run)
saurjya/asteroid/asteroid/masknn/_dcunet_architectures.py
code served (permissive licence) · get_code("a04bf176645c585f")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 0 the same month, 5 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
torch_complex_from_reim 5 other papers
1 dated earlier, 4 later
1910.10400 (2019-10)
Filterbank design for end-to-end speech separation
2008.00264 (2020-08)
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
2010.04228 (2020-10)
All for One and One for All: Improving Music Separation by Bridging Networks
and 2 more not named here
cal_SISNRi 4 other papers
4 dated earlier
1809.07454 (2018-09)
Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
1910.06379 (2019-10)
Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation
1910.14104 (2019-10)
End-to-end Microphone Permutation and Number Invariant Multi-channel Speech Separation
and 1 more not named here
batch_matrix_norm 3 other papers
1 dated earlier, 2 later
1910.10400 (2019-10)
Filterbank design for end-to-end speech separation
2008.00264 (2020-08)
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
2010.04228 (2020-10)
All for One and One for All: Improving Music Separation by Bridging Networks
cal_SISNR 3 other papers
3 dated earlier
1809.07454 (2018-09)
Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
1910.06379 (2019-10)
Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separation
2003.01531 (2020-03)
Voice Separation with an Unknown Number of Multiple Speakers
deep_clustering_loss 3 other papers
1 dated earlier, 2 later
1910.10400 (2019-10)
Filterbank design for end-to-end speech separation
2008.00264 (2020-08)
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
2010.04228 (2020-10)
All for One and One for All: Improving Music Separation by Bridging Networks
is_torch_complex 3 other papers
1 dated earlier, 2 later
1910.10400 (2019-10)
Filterbank design for end-to-end speech separation
2008.00264 (2020-08)
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
2010.04228 (2020-10)
All for One and One for All: Improving Music Separation by Bridging Networks
torch_complex_from_magphase 3 other papers
1 dated earlier, 2 later
1910.10400 (2019-10)
Filterbank design for end-to-end speech separation
2008.00264 (2020-08)
DCCRN: Deep Complex Convolution Recurrent Network for Phase-Aware Speech Enhancement
2010.04228 (2020-10)
All for One and One for All: Improving Music Separation by Bridging Networks

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

The dominant speech separation models are based on complex recurrent or convolution neural network that model speech sequences indirectly conditioning on context, such as passing information through many intermediate states in recurrent neural network, leading to suboptimal separation performance. In this paper, we propose a dual-path transformer network (DPTNet) for end-to-end speech separation, which introduces direct context-awareness in the modeling for speech sequences. By introduces a improved transformer, elements in speech sequences can interact directly, which enables DPTNet can model for the speech sequences with direct context-awareness. The improved transformer in our approach learns the order information of the speech sequences without positional encodings by incorporating a recurrent neural network into the original transformer. In addition, the structure of dual paths makes our model efficient for extremely long speech sequence modeling. Extensive experiments on benchmark datasets show that our approach outperforms the current state-of-the-arts (20.6 dB SDR on the public WSj0-2mix data corpus).

For agents

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

get_harvested_code_for_paper("2007.13975")
get_code_for_paper("2007.13975")
have("2007.13975")

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

curl https://syntology.ai/api/ran/2007.13975.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.13975.svg)](https://syntology.ai/paper/2007.13975)

Connect an agent — have() is free.