Papers › Weakly-supervised Audio Separation via Bi-modal Semantic Similarity

Weakly-supervised Audio Separation via Bi-modal Semantic Similarity

2 Apr 2024arXiv:2404.01740archive 2025-07-28

Tanvir Mahmud, Saeed Amizadeh, Kazuhito Koishida, Diana Marculescu

Conditional sound separation in multi-source audio mixtures without having access to single source sound data during training is a long standing challenge. Existing mix-and-separate based methods suffer from significant performance drop with multi-source training mixtures due to the lack of supervision signal for single source separation cases during training. However, in the case of language-conditional audio separation, we do have access to corresponding text descriptions for each audio mixture in our training data, which can be seen as (rough) representations of the audio samples in the language modality. To this end, in this paper, we propose a generic bi-modal separation framework which can enhance the existing unsupervised frameworks to separate single-source signals in a target modality (i.e., audio) using the easily separable corresponding signals in the conditioning modality (i.e., language), without having access to single-source samples in the target modality during training. We empirically show that this is well within reach if we have access to a pretrained joint embedding model between the two modalities (i.e., CLAP). Furthermore, we propose to incorporate our framework into two fundamental scenarios to enhance separation performance. First, we show that our proposed methodology significantly improves the performance of purely unsupervised baselines by reducing the distribution shift between training and test samples. In particular, we show that our framework can achieve 71% boost in terms of Signal-to-Distortion Ratio (SDR) over the baseline, reaching 97.5% of the supervised learning performance. Second, we show that we can further improve the performance of the supervised learning itself by 17% if we augment it by our proposed weakly-supervised framework, that enables a powerful semi-supervised framework for audio separation.

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: 12 ran (of which 7 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 18 samples from 1 repository, 12 ran (11 with no instrument failure; 1 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 2026-09-27 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

halvani/constituent-treelib officialmentioned in papermentioned on GitHubpytorchMITSyntology: not harvested report
microsoft/bimodalaudioseparation officialmentioned in paperpytorchMITSyntology: 18 samples, 12 ran (1 instrument-failed) · 6 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: 12 ran (of which 7 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 1 where Syntology's instrument failed) · 6 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
1ran · our draft was wrong
10ran
6unverified

Licence: 2 of the 18 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from microsoft/bimodalaudioseparation. “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 3 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.

microsoft/bimodalaudioseparation official repository 18 samples, 12 ran (1 instrument-failed) · 6 unverified
BlockCA models/cond_unet_attn.py ran MIT (permissive) · 84919ee649260509 · report fixture: returned BlockCA · level recorded 2026-09-22
BlockCat models/cond_unet_attn.py ran MIT (permissive) · b733542e13e4b8ce · report fixture: returned BlockCat · level recorded 2026-09-22
BlockIdentity models/cond_unet_attn.py ran fingerprinted MIT (permissive) · 8ec43e0a25ffe65f · report fixture: returned BlockIdentity · level recorded 2026-09-22
BlockSA models/cond_unet_attn.py ran MIT (permissive) · 903f2610f1453cea · report fixture: returned BlockSA · level recorded 2026-09-22
CrossAttention in 85 papers models/cond_unet_attn.py ran MIT (permissive) · 837dd2dcaeaaded4 · report fixture: returned CrossAttention · level recorded 2026-09-22
Mlp in 102 papers models/cond_unet_attn.py ran MIT (permissive) · 033bf6b0228555c7 · report fixture: returned Mlp · 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
models/cond_unet_attn.py ran MIT (permissive) · a84641b3af546ce0 · report fixture: returned ResBlock · level recorded 2026-09-22
SelfAttention in 66 papers models/cond_unet_attn.py ran MIT (permissive) · fee297065b6d46a5 · report fixture: signature x:Tensor/3/float · returned Tensor · level recorded 2026-09-24
load_state_dict in 114 papers models/laion_clap/clap_module/factory.py ran MIT (permissive) here; the stored copy is pointer only · af72484d5cfd7097 · report fixture: signature checkpoint_path:str/0/str, map_location:str/0/str, skip_params:bool/0/bool · returned dict · level recorded 2026-09-24
realimag utils.py ran fingerprinted MIT (permissive) · 81e9a172efc0d908 · report fixture: signature mag:Tensor/2/float, cos:Tensor/2/float, sin:Tensor/2/float · returned tuple · level recorded 2026-09-24
recover_rgb utils.py ran · our draft was wrong MIT (permissive) here; the stored copy is pointer only · 3e5b3b6994917c24 · report fixture: returned ndarray · level recorded 2026-09-27
warpgrid in 6 papers utils.py ran · honoured contract fingerprinted MIT (permissive) · a4a08206de81190b · report fixture: signature bs:int/0/int, HO:int/0/int, WO:int/0/int, warp:bool/0/bool · returned ndarray · level recorded 2026-09-22
BlockSCA models/cond_unet_attn.py unverified MIT (permissive) · da16db8294ed4818 · report
BlockSCACat models/cond_unet_attn.py unverified MIT (permissive) · e4e711f2a5fb5898 · report
CondAttUNet models/cond_unet_attn.py unverified MIT (permissive) · 9811209e675df9fc · report
CondAttUNetBlock models/cond_unet_attn.py unverified MIT (permissive) · bdd91cc0612d6448 · report
create_model in 163 papers models/laion_clap/clap_module/factory.py unverified MIT (permissive) · 5c0daaa41ed490ea · report
create_model_and_transforms in 8 papers models/laion_clap/clap_module/factory.py unverified MIT (permissive) · ffdbb019791354d1 · 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 Bi-modal separation framework (graph method: Bi-modal separation framework). 12 bricks; individually on a fixture: 8 ran (of which 7 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 0 violated, 8 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified. The assembly has not been run end-to-end.

  1. CondAttUNet entry · level 0 · unverified · 9811209e675df9fc
  2. BlockCA model · level 1 · ran · 84919ee649260509
  3. BlockCat helper · level 1 · ran · b733542e13e4b8ce
  4. BlockIdentity helper · level 1 · ran · 8ec43e0a25ffe65f
  5. BlockSA helper · level 1 · ran · 903f2610f1453cea
  6. BlockSCA model · level 0 · unverified · da16db8294ed4818
  7. BlockSCACat model · level 0 · unverified · e4e711f2a5fb5898
  8. CondAttUNetBlock model · level 0 · unverified · bdd91cc0612d6448
  9. CrossAttention model · level 1 · ran · 837dd2dcaeaaded4
  10. Mlp helper · level 1 · ran · 033bf6b0228555c7
  11. SelfAttention model · level 1 · ran · fee297065b6d46a5
  12. ResBlock model · level 1 · ran · a84641b3af546ce0

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="2404.01740") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Semantic SimilaritySemantic Textual Similarity

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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