AV-SSAN: Audio-Visual Selective DoA Estimation through Explicit Multi-Band Semantic-Spatial Alignment

Hongxu Zhu; Jiadong Wang; Kainan Chen; Xinyuan Qian; Yu Chen

REVIEW 6 major objections 5 minor 6 references

AV-SSAN: Audio-Visual Selective DoA Estimation through Explicit Multi-Band Semantic-Spatial Alignment

T0 review · 6 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper proposes that a sound source can be selectively localized from a photo of a different instance of the same class, via explicit semantic-then-spatial alignment, and reports top results on a new benchmark.

desk verdict Genuinely new task and dataset, but the synthetic benchmark's realism is untested and the STARSS23 result doesn't demonstrate transfer; worth refereeing with revisions. read the letter →

arxiv 2507.07384 v2 pith:4C2UA5M3 submitted 2025-07-10 cs.SD eess.AS

Yu Chen , Hongxu Zhu , Jiadong Wang , Kainan Chen , Xinyuan Qian This is my paper

classification cs.SDeess.AS

keywords audio-visualsoundsourcelocalizationdirection-of-arrivalestimationcross-instancevisualpromptingsemantic-spatialalignmentmulti-bandfrequencydecompositionselectiveVGGSound-SSLdataset

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper introduces a task, Cross-Instance Audio-Visual Localization, in which the goal is to estimate the direction of a target sound using only a photo of a different object of the same category, with no requirement that the photo and the sound come from the same recording. It argues that existing audio-visual localization methods fail this task because they align audio and video spatially ('where') without first establishing semantic identity ('what'). To solve it, the paper proposes AV-SSAN, which builds a semantic bridge between the prompt and the audio, then aligns each frequency band of the spatial audio with that semantics and refines the band-wise cues to predict direction of arrival. On the newly constructed VGGSound-SSL benchmark, AV-SSAN reports 16.59 degrees mean absolute error and 71.29 percent accuracy at 0 dB SNR, and it also reports the best transfer result on the real-world STARSS23 set. If these results hold, selective, prompt-driven localization becomes practical without spatially paired audio-visual data.

What carries the argument

The central object is the Multi-Band Semantic-Spatial Alignment Network (MB-SSA Net), a module that couples semantic identity with frequency-dependent spatial cues. It splits each spectrogram into three resolutions—fine bands of 32-bin width, mid bands of 128-bin width, and the full spectrum—and computes interaural phase difference (IPD) and interaural level difference (ILD) per band. Each band passes through a Semantic-Spatial Alignment (SSA) block in which the semantic bridge feature (built from the visual prompt and semantic audio embedding) acts as the query in cross-attention over the band's spatial features; an attention map from the same query-key product yields scalar band-importance weights that aggregate the patches. A spatial refiner then predicts temporal weights across bands and a time-frequency mask that is applied back to the original spectrogram before the DoA predictor. The design carries the argument because it lets the network suppress interference-dominated frequencies and keep target-dominated ones, which the ablations connect to most of the accuracy gain.

What would settle it

Train AV-SSAN only on synthetic VGGSound-SSL, then record a two-microphone array in a real room with a known source position and a same-class prompt image; if the model's MAE on these real recordings is no better than the audio-only baselines, or its accuracy approaches chance, the synthetic spatialization premise fails.

Watch

Extended reading notes

Core claim

The central claim is that selective sound localization can be driven by semantic, not spatial, correspondence: a visual prompt from a different instance of the same class is enough to isolate and localize the matching source inside a mixture of two sounds. The authors implement this as an explicit two-stage pipeline: first, a semantic bridge fuses the prompt's visual embedding with a semantic audio embedding; second, the Multi-Band Semantic-Spatial Alignment Network decomposes the spectrogram into fine, mid, and full frequency bands, computes interaural phase and level differences, and uses the semantic bridge as a query in cross-attention so each band selects target-related spatial evidence. A semantic-guided band fusion weights the bands and a spatial refiner applies the resulting mask to the original spectrogram before DoA prediction. On the synthetic VGGSound-SSL test at 0 dB SNR the framework reaches 16.59 degrees MAE and 71.29 percent accuracy, and on STARSS23 it reaches 27.46 degrees MAE and 52.31 percent accuracy, both ahead of the audio-only and audio-visual baselines compared in the paper.

Load-bearing premise

The central premise is that the synthetic room impulse responses used to spatialize VGGSound audio produce interaural cues and ground-truth directions representative enough of real rooms that a model trained on them will localize in the wild; if that premise fails, the VGGSound-SSL results overstate real localization ability and STARSS23 becomes the only test.

Editorial extensions

If this is right

A user can select which sounding object to locate using only a photo of a different object of the same category, so applications like robotic listening, hearing-assist devices, or video surveillance can query a target without needing the exact paired recording.
Existing audio-visual localization methods, which fuse spatially aligned audio and video, are outperformed by the semantic alignment route on the new benchmark, suggesting that semantic identity should be disentangled from spatial fusion.
Decomposing spatial cues by frequency band and weighting bands by semantic relevance yields gains beyond full-band fusion; the ablation attributes about three percentage points of accuracy to multi-band modeling and band attention.
The VGGSound-SSL dataset (13,981 clips, 296 classes) provides a shared benchmark for cross-instance audio-visual localization, and the reported transfer to STARSS23 indicates the method behaves on real recordings.
Because the visual prompt enters through a semantic embedding rather than a spatially paired frame, the same design could in principle accept other semantic query modalities, such as text descriptions or reference audio, extending selective localization to categories without any image.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Extension beyond the paper: the learned band-attention weights could be read as an explanation of which frequency region carries the target, turning the model into a diagnostic for acoustic cue use, though the paper does not evaluate this interpretation.
Extension beyond the paper: if synthetic spatialization transfers as claimed, large video corpora can be converted into spatial-audio training data without recording arrays, which would make DoA training scalable to far more sound classes than current real-world datasets.
Extension beyond the paper: the paper leaves open how performance degrades when the prompt image is weakly representative of the target class; a stress test with noisy or atypical same-class prompts would clarify the limits of the semantic-spatial alignment.
Extension beyond the paper: the two-stage what-then-where design may apply to other multimodal localization problems, such as tracking a speaker named by a text caption or by a reference voice, where the visual reference is not spatially aligned with the audio.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Jian,

The paper is worth a look, but the headline result is less solid than it appears. The genuinely new thing here is the CI-AVL task — localizing a sound source from a visual prompt of a different instance of the same class — and the VGGSound-SSL dataset built for it. That framing is useful and I don't know a prior paper that does exactly this. The architecture is a reasonable combination of known pieces: CLIP+VGGish for semantics, tri-band IPD/ILD for spatial features, cross-attention, and a mask-based refiner. The ablations show each component adds something, and the qualitative examples are consistent with the story.

The soft spots are real, though. The biggest one: the paper claims VGGSound-SSL is a valuable benchmark, but its realism is never tested. The dataset is made by convolving VGGSound audio with GPU-RIR impulses. All the headline numbers in Table 2 come from that synthetic data. The STARSS23 experiment does not validate the synthetic cues, because the models are retrained on STARSS23 instead of transferring from the synthetic set. The text even calls it a 'transferability' check, which overstates what was done. On top of that, the microphone array geometry used to generate the two-channel audio isn't given in the main text, so IPD/ILD are not reproducible from the description.

There are other issues, mostly in the minor-to-moderate range. The tables have no error bars or significance tests; everything looks like a single run. The baselines are AV-SSL methods that expect spatially-paired video frames, and they're fed cross-instance prompts they weren't designed for, so the comparison is not entirely clean — though it's a reasonable first benchmark. The sigma_theta in the DoA target distribution isn't reported, which matters because the loss is an MSE between predicted and target distributions.

The citation pattern looks fine, and the work is clearly presented. No circular argument or fabricated numbers. The central assumption — that synthetic RIRs give realistic spatial cues — is plausible but unverified.

Who is this for? People working on sound event localization, selective auditory attention, or audio-visual perception will want to know about the task and dataset. It deserves a serious referee, but the authors should be pushed to either demonstrate synthetic-to-real transfer (train on VGGSound-SSL, test on STARSS23 or another real dataset) or soften the claim about the benchmark, and to add variance estimates and release the code and data.

My take: send it to review, but flag the transfer experiment as the key revision.

Referee Report

6 major / 5 minor

Summary. The paper proposes a new task, Cross-Instance Audio-Visual Localization (CI-AVL), in which the direction-of-arrival (DoA) of a target sound is estimated from two-channel audio given a visual prompt from a different instance of the same semantic class. The authors construct VGGSound-SSL, a synthetic dataset of 13,981 two-channel clips created by convolving VGGSound audio with GPU-RIR room impulse responses, and propose AV-SSAN, a model built around a Multi-Band Semantic-Spatial Alignment Network (MB-SSA Net) that decomposes the spectrogram into fine, mid, and full bands, aligns each band with the visual prompt, and refines the spatial cues before DoA prediction. Experiments on VGGSound-SSL at three SNRs and on STARSS23 report that AV-SSAN outperforms eight audio-only and audio-visual baselines, and ablations show monotonic gains from each component.

Significance. If the results hold, the paper contributes a useful task formulation and a plausible architecture. CI-AVL relaxes the spatial-pairing requirement of prior audio-visual sound source localization work, and the explicit multi-band semantic-spatial alignment is an interesting design that is supported by the ablation study. The dataset, if properly validated, could become a large-scale benchmark for prompt-based DoA estimation. However, the current evidence is not yet sufficient: the synthetic benchmark's acoustic realism is unvalidated, the reported gains lack statistical support, several central configuration choices are undocumented, and the comparison set omits the most closely related selective-localization methods.

major comments (6)

[Dataset Construction (Spatial Audio Synthesis)] The array geometry used with GPU-RIR is never specified: the reader is not told the number and spacing of microphones, their orientation, or the source-receiver configuration that produces the two-channel inputs. Since the model's primary cues are IPD and ILD (Eqs. (2)-(3)), which are meaningful only relative to a physical array, this omission blocks reproduction and prevents any check that the simulated spatial cues resemble real recordings. Please report the geometry and include a validation of the RIR simulation against measured array data or a real-recorded benchmark.
[Training and Evaluation; Tables 2 and 3] All reported numbers come from a single training run with no error bars, confidence intervals, or significance tests. The word 'significantly' in the abstract is therefore not supported, and several margins are small: at 0 dB, CMAF (18.65) and AV-SELD (18.95) are close, while on STARSS23 AV-SELD (28.54) and AV-SSAN (27.46) differ by about 1 degree. Please report means and standard deviations over at least three seeds and, where possible, paired significance tests between the best baseline and AV-SSAN.
[Problem Formulation, Eq. (13)] The target distribution p(theta) is written as exp(-|theta_hat - theta|/sigma_theta), but theta_hat is used both as the discretized DoA variable and later as the predicted class, so as written the training target depends on the model's own prediction. Please use distinct notation for the ground-truth class and the predicted class, and report the value of sigma_theta, which is never given and controls the width of the target distribution and hence the scale of L_DoA.
[Training and Evaluation; Table 3] The STARSS23 experiment does not test synthetic-to-real transfer: the text says models are additionally trained on STARSS23, so Table 3 measures performance after training or fine-tuning on real data, not whether cues learned on VGGSound-SSL generalize. The paragraph calling this 'transferability' is misleading. A direct evaluation of a VGGSound-SSL-trained model on STARSS23 would address this; otherwise the wording should be revised and the limitation stated.
[Baselines; Training and Evaluation] The closest prior work in the selective-localization setting (class-conditioned SELD, text-queried SSL, and LocSelect) is discussed in Related Work but never evaluated. Since CI-AVL is a selective localization task, comparisons only against non-selective AV-SSL baselines do not establish the advantage of the proposed approach over methods designed for target-conditioned localization. Please add at least one such comparison, or justify why these methods cannot be adapted to visual prompts.
[Training Loss, Eq. (12)] The ground-truth spectrogram X_gt used in the reconstruction loss is not defined. If X is the mixed spectrogram and X_gt is the target-only spectrogram, then the loss trains a separation mask and this should be stated; if X_gt equals the input, the loss is a trivial identity objective. Please define X_gt precisely and specify whether the mask is supervised by the target source's spectrogram.

minor comments (5)

[Throughout] The dataset name is inconsistent: Table 2's caption and the conclusion call it 'VGG-SSL' while the rest of the paper calls it 'VGGSound-SSL'; please standardize.
[Experiments and Discussions] The phrase 'tested under SNRs ranging from -5 dB to -10 dB' should read 'at -5 dB and -10 dB,' since training is at 0 dB.
[Problem Formulation] p(theta) is described as 'Gaussian-like' but exp(-|x|/sigma) is a Laplace distribution; please correct the terminology.
[MB-SSA Net, Eqs. (6)-(7)] A_b^i has matrix form but is then treated as a vector when concatenated and pooled; please specify the exact dimensions and the pooling operation used to obtain the scalar beta_b^i.
[Figure 4] The color/weight encoding ('Larger Weights', 'Lower Weights') in the figure is not explained in the caption; please add a legend or explicit caption text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the DoA prediction and losses are standard supervised targets; the self-constructed VGGSound-SSL dataset is a benchmark choice, not a circular reduction.

full rationale

The paper's claimed derivation chain is an empirical supervised-learning pipeline, not a derivation from fitted constants. The DoA target p(θ) is a Gaussian-like vector centered on the RIR-simulated ground-truth angle; the model output is a softmax over 180 classes and is trained with an MSE loss against that target (Eq. 13). No equation reduces to its own input: the reconstruction loss (Eq. 12) compares the masked spectrogram to the ground-truth target spectrogram, and the mask itself is predicted from the input mixture, so the training target is not a renamed version of the input. The only self-referential element is that the principal benchmark, VGGSound-SSL, is constructed by the authors via GPU-RIR convolution; this is a dataset-construction choice and not a circular step. The STARSS23 evaluation is external, although the paper phrases it as 'transferability' while actually retraining on STARSS23; this is a wording and validity concern, not circularity. Self-citations to LocSelect and to Qian et al. prior AV-SSL works appear only as related-work context and baselines, not as load-bearing justification for the method's predictions. Therefore no circular step can be quoted from the paper's equations or argument structure.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. The central empirical claim depends on standard spatial-audio assumptions (IPD/ILD sufficiency), a synthetic RIR-based training distribution, and the learnability of cross-modal alignment from pretrained CLIP and VGGish features. Hand-chosen hyperparameters (sigma_theta, band widths, training SNR) are not swept in the main text.

free parameters (3)

sigma_theta = not reported
Standard deviation of the exponential posterior target p(theta) = exp(-|theta_hat - theta| / sigma_theta) in Eq. 13; controls the width of the target distribution and therefore the DoA loss. Value missing from main text.
tri-band widths = fine=32, mid=128, full spectrum
The MB-SSA Net decomposes the spectrogram into three hand-chosen frequency resolutions; these widths are central to the method and no sensitivity analysis is provided in the main text.
training SNR = 0 dB
All models are trained with target and interference mixed at 0 dB SNR; the choice of this level is arbitrary and may influence how well the model handles other SNRs.

assumptions (4)

domain assumption IPD and ILD extracted from STFT are sufficient spatial cues for DoA estimation
Used in Eq. 2-3; standard in SSL, but no proof that these features capture the target's direction in reverberant mixtures.
domain assumption Synthetic spatialization via GPU-RIR produces realistic binaural/spatial audio
Dataset construction convolved VGGSound clips with simulated RIRs; generalization to real recordings is assumed and only partially tested on STARSS23.
ad hoc to paper Semantic alignment between CLIP visual embeddings and VGGish audio embeddings can be learned by a Transformer after linear projection
The design of SBF Net assumes these pretrained embeddings are compatible enough; no theoretical guarantee, and the alignment is trained on the synthetic dataset.
domain assumption The 'what then where' perception hypothesis motivates the two-stage alignment
Cites human perception literature; the architecture encodes this ordering but the hypothesis is not isolated in experiments.

how reviews work

0 comments

Cite this review

Pith. "Pith review of AV-SSAN: Audio-Visual Selective DoA Estimation through Explicit Multi-Band Semantic-Spatial Alignment." pith.science (2026). https://pith.science/paper/4C2UA5M3

@misc{pith2026250707384,
  author       = {Pith},
  title        = {Pith review of: AV-SSAN: Audio-Visual Selective DoA Estimation through Explicit Multi-Band Semantic-Spatial Alignment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4C2UA5M3}},
  note         = {Machine review of arXiv:2507.07384}
}

read the original abstract

Audio-visual sound source localization (AV-SSL) estimates the position of sound sources by fusing auditory and visual cues. Current AV-SSL methodologies typically require spatially-paired audio-visual data and cannot selectively localize specific target sources. To address these limitations, we introduce Cross-Instance Audio-Visual Localization (CI-AVL), a novel task that localizes target sound sources using visual prompts from different instances of the same semantic class. CI-AVL enables selective localization without spatially paired data. To solve this task, we propose AV-SSAN, a semantic-spatial alignment framework centered on a Multi-Band Semantic-Spatial Alignment Network (MB-SSA Net). MB-SSA Net decomposes the audio spectrogram into multiple frequency bands, aligns each band with semantic visual prompts, and refines spatial cues to estimate the direction-of-arrival (DoA). To facilitate this research, we construct VGGSound-SSL, a large-scale dataset comprising 13,981 spatial audio clips across 296 categories, each paired with visual prompts. AV-SSAN achieves a mean absolute error of 16.59 and an accuracy of 71.29%, significantly outperforming existing AV-SSL methods. Code and data will be public.

Figures

Figures reproduced from arXiv: 2507.07384 by the authors.

**Figure 2.** The architecture of (a) our proposed AV-SSAN and (b) Semantic-Spatial Alignment (SSA) block. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Visualizations of AV-SSAN’s outputs on VGG-SSL. Each row shows a pair of overlapping sound events categorized [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Visualizations of the ablation study on band attention. Our model assigns higher weights to frequency bands dominated [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

6 extracted references · 4 canonical work pages

[180]

Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al

Springer. Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al. 2020. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929. Gulati, A.; Qin, J.; Chiu, C.-C.; Parmar, N.; Zhang, Y .; Yu, J.; Han, W.; Wang, S.; Z...

arXiv 2020
[2010]

van der Heijden, K.; Rauschecker, J

Semantic elaboration in auditory and visual spatial memory.Frontiers in Psychology, 1: 228. van der Heijden, K.; Rauschecker, J. P.; de Gelder, B.; and Formisano, E. 2019. Cortical mechanisms of spatial hearing. Nature Reviews Neuroscience, 20(10): 609–623. Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A. N.; Kaiser, Ł.; and Polos...

arXiv 2019
[2018]

IEEE Journal of Selected Topics in Signal Processing, 13(1): 34–48

Sound event localization and detection of overlap- ping sources using convolutional recurrent neural networks. IEEE Journal of Selected Topics in Signal Processing, 13(1): 34–48. Berghi, D.; Wu, P.; Zhao, J.; Wang, W.; and Jackson, P. J

work page
[2022]

InIEEE International Conference on Acous- tics, Speech and Signal Processing, 711–715

Locate this, not that: Class-conditioned sound event doa estimation. InIEEE International Conference on Acous- tics, Speech and Signal Processing, 711–715. IEEE. Strutt, J. W. 1907. On our perception of sound direction. Philosophical Magazine, 13(74): 214–32. Taevs, M.; Dahmani, L.; Zatorre, R. J.; and Bohbot, V . D

work page 1907
[2024]

InIEEE International Con- ference on Acoustics, Speech and Signal Processing, 8816–

Fusion of audio and visual embeddings for sound event localization and detection. InIEEE International Con- ference on Acoustics, Speech and Signal Processing, 8816–

work page
[8820]

Brughera, A.; Dunai, L.; and Hartmann, W

IEEE. Brughera, A.; Dunai, L.; and Hartmann, W. M. 2013. Hu- man interaural time difference thresholds for sine tones: The high-frequency limit.The Journal of the Acoustical Society of America, 133(5): 2839–2855. Chakrabarty, S.; and Habets, E. A. 2017. Broadband DOA estimation using convolutional neural networks trained with noise signals. InIEEE Worksho...

work page 2013

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [180]

Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al

Springer. Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al. 2020. An image is worth 16x16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929. Gulati, A.; Qin, J.; Chiu, C.-C.; Parmar, N.; Zhang, Y .; Yu, J.; Han, W.; Wang, S.; Z...

arXiv 2020

[2] [2010]

van der Heijden, K.; Rauschecker, J

Semantic elaboration in auditory and visual spatial memory.Frontiers in Psychology, 1: 228. van der Heijden, K.; Rauschecker, J. P.; de Gelder, B.; and Formisano, E. 2019. Cortical mechanisms of spatial hearing. Nature Reviews Neuroscience, 20(10): 609–623. Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A. N.; Kaiser, Ł.; and Polos...

arXiv 2019

[3] [2018]

IEEE Journal of Selected Topics in Signal Processing, 13(1): 34–48

Sound event localization and detection of overlap- ping sources using convolutional recurrent neural networks. IEEE Journal of Selected Topics in Signal Processing, 13(1): 34–48. Berghi, D.; Wu, P.; Zhao, J.; Wang, W.; and Jackson, P. J

work page

[4] [2022]

InIEEE International Conference on Acous- tics, Speech and Signal Processing, 711–715

Locate this, not that: Class-conditioned sound event doa estimation. InIEEE International Conference on Acous- tics, Speech and Signal Processing, 711–715. IEEE. Strutt, J. W. 1907. On our perception of sound direction. Philosophical Magazine, 13(74): 214–32. Taevs, M.; Dahmani, L.; Zatorre, R. J.; and Bohbot, V . D

work page 1907

[5] [2024]

InIEEE International Con- ference on Acoustics, Speech and Signal Processing, 8816–

Fusion of audio and visual embeddings for sound event localization and detection. InIEEE International Con- ference on Acoustics, Speech and Signal Processing, 8816–

work page

[6] [8820]

Brughera, A.; Dunai, L.; and Hartmann, W

IEEE. Brughera, A.; Dunai, L.; and Hartmann, W. M. 2013. Hu- man interaural time difference thresholds for sine tones: The high-frequency limit.The Journal of the Acoustical Society of America, 133(5): 2839–2855. Chakrabarty, S.; and Habets, E. A. 2017. Broadband DOA estimation using convolutional neural networks trained with noise signals. InIEEE Worksho...

work page 2013