Papers › An efficient encoder-decoder architecture with top-down attention for speech separation

An efficient encoder-decoder architecture with top-down attention for speech separation

30 Sep 2022arXiv:2209.15200archive 2025-07-28

Kai Li, Runxuan Yang, Xiaolin Hu

Deep neural networks have shown excellent prospects in speech separation tasks. However, obtaining good results while keeping a low model complexity remains challenging in real-world applications. In this paper, we provide a bio-inspired efficient encoder-decoder architecture by mimicking the brain's top-down attention, called TDANet, with decreased model complexity without sacrificing performance. The top-down attention in TDANet is extracted by the global attention (GA) module and the cascaded local attention (LA) layers. The GA module takes multi-scale acoustic features as input to extract global attention signal, which then modulates features of different scales by direct top-down connections. The LA layers use features of adjacent layers as input to extract the local attention signal, which is used to modulate the lateral input in a top-down manner. On three benchmark datasets, TDANet consistently achieved competitive separation performance to previous state-of-the-art (SOTA) methods with higher efficiency. Specifically, TDANet's multiply-accumulate operations (MACs) are only 5% of Sepformer, one of the previous SOTA models, and CPU inference time is only 10% of Sepformer. In addition, a large-size version of TDANet obtained SOTA results on three datasets, with MACs still only 10% of Sepformer and the CPU inference time only 24% of Sepformer.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 18 code samples harvested from 1 repository linked to this paper: 13 ran (of which 8 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; 1 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: official repository: 18 samples from 1 repository, 13 ran (9 with no instrument failure; 4 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 13 samples that ran has no recorded date.

JusperLee/TDANet officialpytorchSyntology: 18 samples, 13 ran (4 instrument-failed) · 5 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

18 samples harvested: 13 ran (of which 8 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

3ran · our draft was wrong
1ran; the contract check could not run
9ran
5unverified

Licence: 0 of the 18 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from JusperLee/TDANet. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 5 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

JusperLee/TDANet official repository 18 samples, 13 ran (4 instrument-failed) · 5 unverified
BaseModel in 92 papers look2hear/models/TDANet.py ran Apache-2.0 (permissive) · 7432d190e3e61daf · report fixture: returned BaseModel · level recorded 2026-09-22
ConvNorm look2hear/models/TDANet.py ran fingerprinted Apache-2.0 (permissive) · a4c13d6751fb8ee9 · report fixture: returned ConvNorm · level recorded 2026-09-22
ConvNormAct look2hear/models/TDANet.py ran fingerprinted Apache-2.0 (permissive) · c92dfb01b6f67601 · report fixture: returned ConvNormAct · level recorded 2026-09-22
DilatedConvNorm look2hear/models/TDANet.py ran fingerprinted Apache-2.0 (permissive) · 7dc0123d9144bdfc · report fixture: returned DilatedConvNorm · level recorded 2026-09-22
GlobLN look2hear/models/TDANet.py ran · our draft was wrong Apache-2.0 (permissive) · a5d6bac3ebd586de · report fixture: returned GroupNorm · level recorded 2026-09-22
GlobalAttention look2hear/models/TDANet.py ran Apache-2.0 (permissive) · 61234fa9cacd6a23 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
LA look2hear/models/TDANet.py ran Apache-2.0 (permissive) · 73ef3a7889f6403a · report fixture: returned LA · level recorded 2026-09-22
Mlp in 102 papers look2hear/models/TDANet.py ran Apache-2.0 (permissive) · c3b7c4dad2d1de89 · report fixture: returned Mlp · level recorded 2026-09-22
MultiHeadAttention in 124 papers
census: 9 implementations from 7 papers, 9 distinct outputs across 3 buckets, one shared input per bucket
look2hear/models/TDANet.py ran fingerprinted Apache-2.0 (permissive) · 62e9ea75f34de78d · report fixture: returned MultiHeadAttention · level recorded 2026-09-22
PositionalEncoding in 126 papers
census: 13 implementations from 9 papers, 9 distinct outputs across 5 buckets, one shared input per bucket
look2hear/models/TDANet.py ran Apache-2.0 (permissive) · b94f4f291747d5f4 · report fixture: returned PositionalEncoding · level recorded 2026-09-22
cached_download look2hear/models/TDANet.py ran · our draft was wrong Apache-2.0 (permissive) · 0a5e1308766cf3ed · report fixture: returned str · level recorded 2026-09-22
drop_path in 325 papers
census: 27 implementations from 61 papers, 5 distinct outputs across 2 buckets, one shared input per bucket this sample: bucket 2 of 2, arg 1: rank 2, kind float, dtype float32
look2hear/models/TDANet.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted Apache-2.0 (permissive) · 04e8544d3f2fd61b · report fixture: signature x:Tensor/2/float, drop_prob:float/0/float, training:bool/0/bool · returned Tensor · level recorded, date not recorded
url_to_filename in 73 papers look2hear/models/TDANet.py ran · our draft was wrong Apache-2.0 (permissive) · 1a28ae0ed1f639b5 · report fixture: returned str · level recorded 2026-09-22
GlobLN look2hear/models/TDANet.py unverified Apache-2.0 (permissive) · 75f33a42561b8c51 · report
Recurrent in 5 papers look2hear/models/TDANet.py unverified Apache-2.0 (permissive) · b122a90008dae0d5 · report
TDANet look2hear/models/TDANet.py unverified Apache-2.0 (permissive) · db5ae3a50a143f24 · report
UConvBlock look2hear/models/TDANet.py unverified Apache-2.0 (permissive) · 9f7cf1cecbacaaba · report
get_cache_dir in 14 papers look2hear/models/TDANet.py unverified Apache-2.0 (permissive) · 52b1562ec0936b6e · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for TDANet (graph method: TDANet). 18 bricks; individually on a fixture: 14 ran (of which 9 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

  1. TDANet entry · level 0 · unverified · db5ae3a50a143f24
  2. BaseModel helper · level 1 · ran · 7432d190e3e61daf
  3. GlobLN helper · level 1 · ran · our draft was wrong · a5d6bac3ebd586de
  4. Recurrent model · level 0 · unverified · b122a90008dae0d5
  5. cached_download helper · level 1 · ran · our draft was wrong · 0a5e1308766cf3ed
  6. UConvBlock model · level 0 · unverified · 9f7cf1cecbacaaba
  7. get_cache_dir helper · level 0 · unverified · 52b1562ec0936b6e
  8. url_to_filename helper · level 1 · ran · our draft was wrong · 1a28ae0ed1f639b5
  9. ConvNormAct helper · level 1 · ran · c92dfb01b6f67601
  10. DilatedConvNorm helper · level 1 · ran · 7dc0123d9144bdfc
  11. GlobalAttention model · level 1 · ran · 61234fa9cacd6a23
  12. LA helper · level 1 · ran · 73ef3a7889f6403a
  13. DropPath helper · level 1 · ran · ad5f80b54573c0e0
  14. Mlp model · level 1 · ran · c3b7c4dad2d1de89
  15. MultiHeadAttention model · level 1 · ran · 62e9ea75f34de78d
  16. drop_path helper · level 1 · ran; the contract check could not run · 55120f2026b56aa2
  17. ConvNorm helper · level 1 · ran · a4c13d6751fb8ee9
  18. PositionalEncoding helper · level 1 · ran · b94f4f291747d5f4

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2209.15200") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

DecoderSpeech Separation

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Speech Separation Libri2Mix TDANet Large SI-SDRi 17.4 #6 of 10 Archive leaderboard report
Speech Separation Libri2Mix TDANet SI-SDRi 16.9 #7 of 10 Archive leaderboard report
Speech Separation WHAM! TDANet Large SI-SDRi 15.2 #4 of 6 Archive leaderboard report
Speech Separation WHAM! TDANet SI-SDRi 14.8 #5 of 6 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

GA

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections