Papers › M3ANet: Multi-scale and Multi-Modal Alignment Network for Brain-Assisted Target...

M3ANet: Multi-scale and Multi-Modal Alignment Network for Brain-Assisted Target Speaker Extraction

31 May 2025arXiv:2506.00466archive 2025-07-28

Cunhang Fan, Ying Chen, Jian Zhou, Zexu Pan, Jingjing Zhang, Youdian Gao, Xiaoke Yang, Zhengqi Wen, Zhao Lv

The brain-assisted target speaker extraction (TSE) aims to extract the attended speech from mixed speech by utilizing the brain neural activities, for example Electroencephalography (EEG). However, existing models overlook the issue of temporal misalignment between speech and EEG modalities, which hampers TSE performance. In addition, the speech encoder in current models typically uses basic temporal operations (e.g., one-dimensional convolution), which are unable to effectively extract target speaker information. To address these issues, this paper proposes a multi-scale and multi-modal alignment network (M3ANet) for brain-assisted TSE. Specifically, to eliminate the temporal inconsistency between EEG and speech modalities, the modal alignment module that uses a contrastive learning strategy is applied to align the temporal features of both modalities. Additionally, to fully extract speech information, multi-scale convolutions with GroupMamba modules are used as the speech encoder, which scans speech features at each scale from different directions, enabling the model to capture deep sequence information. Experimental results on three publicly available datasets show that the proposed model outperforms current state-of-the-art methods across various evaluation metrics, highlighting the effectiveness of our proposed method. The source code is available at: https://github.com/fchest/M3ANet.

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 41 code samples harvested from 1 repository linked to this paper: 30 ran (of which 23 constructed an object rather than computing a result; 28 with no instrument failure: 0 honoured, 0 violated, 28 with no contract checked; 2 where Syntology's instrument failed) · 11 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 41 samples from 1 repository, 30 ran (28 with no instrument failure; 2 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.

fchest/m3anet officialmentioned in paperpytorchSyntology: 41 samples, 30 ran (2 instrument-failed) · 11 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

41 samples harvested: 30 ran (of which 23 constructed an object rather than computing a result; 28 with no instrument failure: 0 honoured, 0 violated, 28 with no contract checked; 2 where Syntology's instrument failed) · 11 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.

2ran · our draft was wrong
28ran
11unverified

Licence: 41 of the 41 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 fchest/m3anet. “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 7 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.

fchest/m3anet official repository 41 samples, 30 ran (2 instrument-failed) · 11 unverified
ChannelwiseLayerNorm M3ANET.py ran fingerprinted no licence file found · pointer only · 735ccb7191f175dd · report fixture: returned ChannelwiseLayerNorm · level recorded 2026-09-22
Chebynet M3ANET.py ran no licence file found · pointer only · 19ef1648b39c176b · report fixture: returned Chebynet · level recorded 2026-09-22
Conv1D in 9 papers M3ANET.py ran fingerprinted no licence file found · pointer only · 9bcc5aec40c941f8 · report fixture: returned Conv1D · level recorded 2026-09-22
ConvTrans1D M3ANET.py ran fingerprinted no licence file found · pointer only · 990690b6a9274151 · report fixture: returned ConvTrans1D · level recorded 2026-09-22
CrossMerge M3ANET.py ran no licence file found · pointer only · 641b132762976941 · report fixture: returned CrossMerge · level recorded 2026-09-22
CrossMerge_1 M3ANET.py ran no licence file found · pointer only · 58436fa87884aea9 · report fixture: returned CrossMerge_1 · level recorded 2026-09-22
CrossMerge_2 M3ANET.py ran no licence file found · pointer only · f20923d150378f69 · report fixture: returned CrossMerge_2 · level recorded 2026-09-22
CrossMerge_3 M3ANET.py ran no licence file found · pointer only · f7d481ff69e34b5b · report fixture: returned CrossMerge_3 · level recorded 2026-09-22
CrossMerge_4 M3ANET.py ran no licence file found · pointer only · 826ffd61623145b6 · report fixture: returned CrossMerge_4 · level recorded 2026-09-22
CrossScan_1 M3ANET.py ran no licence file found · pointer only · 4b0f4e0cae52a70e · report fixture: returned CrossScan_1 · level recorded 2026-09-22
CrossScan_2 M3ANET.py ran no licence file found · pointer only · 90fc0ee8b63169e7 · report fixture: returned CrossScan_2 · level recorded 2026-09-22
CrossScan_3 M3ANET.py ran no licence file found · pointer only · 82a5934402ec929a · report fixture: returned CrossScan_3 · level recorded 2026-09-22
CrossScan_4 M3ANET.py ran no licence file found · pointer only · a60ee2d06c793043 · report fixture: returned CrossScan_4 · level recorded 2026-09-22
DepthConv1d M3ANET.py ran fingerprinted no licence file found · pointer only · 0c6fb0ea81f502ff · report fixture: returned DepthConv1d · level recorded 2026-09-22
Dual_RNN_Block M3ANET.py ran fingerprinted no licence file found · pointer only · cf54607aca7342a2 · report fixture: returned Dual_RNN_Block · level recorded 2026-09-22
EEGEncoder M3ANET.py ran no licence file found · pointer only · 623f136c1c30728c · report fixture: returned EEGEncoder · level recorded 2026-09-22
GridNetBlock M3ANET.py ran no licence file found · pointer only · 08fa276362859263 · report fixture: returned GridNetBlock · level recorded 2026-09-22
LayerNorm2d in 18 papers M3ANET.py ran fingerprinted no licence file found · pointer only · 55a4c3d91ff8953f · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
LayerNormalization4D M3ANET.py ran fingerprinted no licence file found · pointer only · ff31689965b652db · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
LayerNormalization4DCF M3ANET.py ran no licence file found · pointer only · 02ffad1ac29ab5c6 · report fixture: returned LayerNormalization4DCF · level recorded 2026-09-22
Linear in 127 papers M3ANET.py ran fingerprinted no licence file found · pointer only · 587d5a31c0d32a3e · report fixture: returned Linear · level recorded 2026-09-22
Linear2d M3ANET.py ran no licence file found · pointer only · ebccefe76531b2fb · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
PVT2FFN M3ANET.py ran no licence file found · pointer only · a9041b3253ce9575 · report fixture: returned PVT2FFN · level recorded 2026-09-22
ResBlock in 127 papers
census: 6 implementations from 5 papers, 6 distinct outputs across 2 buckets, one shared input per bucket
M3ANET.py ran fingerprinted no licence file found · pointer only · 013eb7e4aa4c04bb · report fixture: returned ResBlock · level recorded 2026-09-22
ScaledDotProductAttention in 46 papers M3ANET.py ran · metamorphic tier: deterministic no licence file found · pointer only · 8ea2b63c43b3af74 · report fixture: signature q:Tensor/4/float, k:Tensor/4/float, v:Tensor/4/float, mask:Tensor/4/float · returned tuple · level recorded 2026-09-22
SelectiveScanCore M3ANET.py ran no licence file found · pointer only · c76558d2228cc5bf · report fixture: returned SelectiveScanCore · level recorded 2026-09-22
SelectiveScanOflex M3ANET.py ran no licence file found · pointer only · 914d56bf38505632 · report fixture: returned SelectiveScanOflex · level recorded 2026-09-22
SoftmaxSpatial M3ANET.py ran fingerprinted no licence file found · pointer only · e0fccfbd47b18704 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
generate_cheby_adj M3ANET.py ran · our draft was wrong no licence file found · pointer only · 79dbcb5c4d2bafa4 · report fixture: returned list · level recorded 2026-09-22
normalize_A M3ANET.py ran · our draft was wrong no licence file found · pointer only · 72fa821cbb7c5ee2 · report fixture: returned Tensor · level recorded 2026-09-22
ConvCrossAttention M3ANET.py unverified no licence file found · pointer only · 8e053d902a16be5b · report
DPRNN M3ANET.py unverified no licence file found · pointer only · 350faceb54d0de7c · report
GraphConvolution in 36 papers M3ANET.py unverified no licence file found · pointer only · e84e06c700e083fc · report
M3ANET M3ANET.py unverified no licence file found · pointer only · de706e120acf2053 · report
MultiLayerCrossAttention M3ANET.py unverified no licence file found · pointer only · 51cee1c75c8a0f31 · report
Permute in 5 papers M3ANET.py unverified no licence file found · pointer only · b20107995853e0bf · report
SS2D M3ANET.py unverified no licence file found · pointer only · 041ba92d7d084dfa · report
SS2Dv2 M3ANET.py unverified no licence file found · pointer only · e4c64428e7712b5a · report
TFGridNet M3ANET.py unverified no licence file found · pointer only · 3d6229eae61bd863 · report
VSSS M3ANET.py unverified no licence file found · pointer only · bb35610b16cdc7e6 · report
mamba_init M3ANET.py unverified no licence file found · pointer only · 9483a14d1d95cf23 · 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 M3ANet (graph method: M3ANet). 43 bricks; individually on a fixture: 32 ran (of which 25 constructed an object rather than computing a result; 30 with no instrument failure: 0 honoured, 0 violated, 30 with no contract checked; 2 where Syntology's instrument failed) · 11 unverified. The assembly has not been run end-to-end.

  1. M3ANET entry · level 0 · unverified · de706e120acf2053
  2. ChannelwiseLayerNorm helper · level 1 · ran · 735ccb7191f175dd
  3. Conv1D helper · level 1 · ran · 9bcc5aec40c941f8
  4. ConvTrans1D helper · level 1 · ran · 990690b6a9274151
  5. DPRNN helper · level 0 · unverified · 350faceb54d0de7c
  6. EEGEncoder model · level 1 · ran · 623f136c1c30728c
  7. TFGridNet helper · level 0 · unverified · 3d6229eae61bd863
  8. VSSS model · level 0 · unverified · bb35610b16cdc7e6
  9. Dual_RNN_Block helper · level 1 · ran · cf54607aca7342a2
  10. MultiLayerCrossAttention helper · level 0 · unverified · 51cee1c75c8a0f31
  11. Chebynet model · level 1 · ran · 19ef1648b39c176b
  12. ResBlock helper · level 1 · ran · 013eb7e4aa4c04bb
  13. normalize_A helper · level 1 · ran · our draft was wrong · 72fa821cbb7c5ee2
  14. GridNetBlock helper · level 1 · ran · 08fa276362859263
  15. CrossMerge_1 helper · level 1 · ran · 58436fa87884aea9
  16. CrossMerge_2 helper · level 1 · ran · f20923d150378f69
  17. CrossMerge_3 helper · level 1 · ran · f7d481ff69e34b5b
  18. CrossMerge_4 helper · level 1 · ran · 826ffd61623145b6
  19. CrossScan_1 helper · level 1 · ran · 4b0f4e0cae52a70e
  20. CrossScan_2 helper · level 1 · ran · 90fc0ee8b63169e7
  21. CrossScan_3 helper · level 1 · ran · 82a5934402ec929a
  22. CrossScan_4 helper · level 1 · ran · a60ee2d06c793043
  23. PVT2FFN helper · level 1 · ran · a9041b3253ce9575
  24. SS2D helper · level 0 · unverified · 041ba92d7d084dfa
  25. ConvCrossAttention helper · level 0 · unverified · 8e053d902a16be5b
  26. GraphConvolution helper · level 0 · unverified · e84e06c700e083fc
  27. generate_cheby_adj helper · level 1 · ran · our draft was wrong · 79dbcb5c4d2bafa4
  28. LayerNormalization4D helper · level 1 · ran · ff31689965b652db
  29. LayerNormalization4DCF helper · level 1 · ran · 02ffad1ac29ab5c6
  30. DWConv helper · level 1 · ran · e68b276a44902fe9
  31. SS2Dv2 helper · level 0 · unverified · e4c64428e7712b5a
  32. mamba_init helper · level 0 · unverified · 9483a14d1d95cf23
  33. DepthConv1d helper · level 1 · ran · 0c6fb0ea81f502ff
  34. ScaledDotProductAttention helper · level 1 · ran · 8ea2b63c43b3af74
  35. CrossMerge helper · level 1 · ran · 641b132762976941
  36. CrossScan helper · level 1 · ran · 930e2c172dd437df
  37. LayerNorm2d helper · level 1 · ran · 55a4c3d91ff8953f
  38. Linear helper · level 1 · ran · 587d5a31c0d32a3e
  39. Linear2d helper · level 1 · ran · ebccefe76531b2fb
  40. Permute helper · level 0 · unverified · b20107995853e0bf
  41. SelectiveScanCore helper · level 1 · ran · c76558d2228cc5bf
  42. SelectiveScanOflex helper · level 1 · ran · 914d56bf38505632
  43. SoftmaxSpatial helper · level 1 · ran · e0fccfbd47b18704

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="2506.00466") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Contrastive LearningEEGTarget Speaker Extraction

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ALIGNContrastive Learning

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