ReverbMiipher: Generative Speech Restoration meets Reverberation Characteristics Controllability

Adriana Guevara-Rukoz; Haruko Ishikawa; Heiga Zen; Michiel Bacchiani; Robin Scheibler; Shigeki Karita; Wataru Nakata; Yuma Koizumi

REVIEW 3 major objections 5 minor 59 references

ReverbMiipher: Generative Speech Restoration meets Reverberation Characteristics Controllability

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper claims that a generative speech-restoration model can remove noise and codec artifacts while preserving the original reverberation, and that it does so by routing a dedicated reverb feature through the vocoder, outperforming the…

desk verdict ReverbMiipher is a solid incremental extension of Miipher-2 that beats a two-stage RIR baseline at preserving reverberation, but the claimed disentanglement of the reverb feature is asserted, not demonstrated. read the letter →

arxiv 2505.05077 v2 pith:BQO7R24T submitted 2025-05-08 cs.SD eess.AS

Wataru Nakata , Yuma Koizumi , Shigeki Karita , Robin Scheibler , Haruko Ishikawa , Adriana Guevara-Rukoz , Heiga Zen , Michiel Bacchiani This is my paper · ORCID

classification cs.SDeess.AS

keywords speechrestorationreverberationpreservationcontrolparametricresynthesisreverbfeaturedisentanglementneuralvocoderlatentspaceinterpolationroomimpulseresponse

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

Speech-restoration systems typically erase the reverberation that tells listeners about the room the speaker was in. ReverbMiipher extends the Miipher-2 parametric-resynthesis architecture with a ReverbEncoder that reads a log-mel spectrogram of the noisy input and produces a 512-dimensional reverb feature; this feature conditions the WaveFit vocoder together with the restored speech features, so the model learns to map the noisy signal to the reverberant clean signal s*r rather than to anechoic speech. A stochastic zero-vector replacement during training teaches the model that a zero vector means 'no reverb', which is what pushes the encoder to encode only reverberation. The paper reports that this one-stage model preserves reverberation better, by subjective ranking and by objective metrics, than restoring clean speech and then convolving a simulated room impulse response, and that interpolation or PCA-plane sampling of the reverb feature yields continuous control over the acoustic environment.

What carries the argument

The central object is the 512-dimensional reverb feature c = R(x) produced by the ReverbEncoder, a stack of four Conformer layers with kernel size five followed by time-axis mean pooling, operating on the same 128-band log-mel input as the feature cleaner. This vector conditions the WaveFit vocoder as V(ĥ_s, c), and the training loss of Eq. (1) switches stochastically with probability q = 0.1 between the reverberant target s*r (when c is the true encoder output) and the anechoic target s (when c is a zero vector). The switching is what forces the encoder to specialize: the vocoder must learn to use c to reproduce the room, while the zero-vector case tells it that no reverb is also a valid condition, so any non-reverb information in c would have to be ignored or would corrupt the anechoic output. That mechanism is what separates the reverb representation from the speech content and makes downstream interpolation and sampling meaningful.

What would settle it

Take a fixed clean anechoic utterance s, extract reverb features c_i from a set of noisy utterances with different speakers and content, and synthesize V(ĥ_s, c_i). If the reverb feature is truly disentangled, the output should have the same ASR transcript and near-identical speaker embedding as s for every c_i, with only the room acoustics changing; a systematic drop in speaker similarity or rise in word error rate as c_i varies would falsify the disentanglement claim.

Watch

Extended reading notes

Core claim

ReverbMiipher's central claim is that reverberation can be treated as an explicit conditioning variable inside a generative speech-restoration model rather than something to remove or add afterwards. The ReverbEncoder's feature vector c is the carrier of the room acoustics: when c is present, the vocoder is trained to output s*r, the clean speech convolved with the original room impulse response, and when c is the zero vector, it must output the anechoic source s. Because the model receives only the noisy input, the encoder must pull the reverb information out of the degraded mixture, and the random switching with probability q = 0.1 is the training signal that prevents c from smuggling in speaker identity, content, or noise. The paper then shows that c varies smoothly with RT60 and DRR in a PCA projection, that linear interpolation between two c vectors produces intermediate reverberation, and that sampling from the PCA plane gives new but structurally intact reverberation effects. The claim is that this one-stage, feature-conditioned design outperforms the two-stage baseline of Miipher-2 restoration followed by convolution with a simulated RIR chosen with known RT60 and DRR.

Load-bearing premise

The whole control scheme rests on the assumption that training with a 10% zero-vector switching probability makes the reverb feature encode only reverberation; if that vector also leaks speaker identity or content, the demonstrated interpolation and sampling are changing the speech itself rather than just the room.

Editorial extensions

If this is right

Reverb preservation becomes a built-in behavior of speech restoration rather than a post-hoc step, so the acoustic environment of the original recording is retained in applications such as augmented reality, virtual reality, and video generation.
The reverb feature can be transferred between utterances, interpolated, or sampled from a PCA plane, enabling the creation of datasets with controllable spatial characteristics.
Because the model is trained end-to-end rather than relying on explicit RIR estimation, it avoids the error propagation and fidelity losses of the two-stage clean-then-convolve pipeline.
The approach beats the simulated-RIR baseline even when that baseline is given perfect knowledge of RT60 and DRR, indicating that the advantage comes from the architecture itself rather than from having better room information.

Reading between the lines

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

If the disentanglement holds, the same conditioning trick could be extended to other acoustic attributes, such as environmental sounds, effectively turning speech restoration into a controllable acoustic-scene remixer.
A testable extension is cross-speaker or cross-language reverb transfer: swapping the reverb feature between utterances with different speakers or content should leave the transcript and speaker identity untouched, which can be verified with ASR and speaker-embedding similarity.
The smooth PCA-plane structure suggests the reverb feature manifold may be low-dimensional; an explicit dimensionality analysis or a learned controllable latent could give even finer control than linear interpolation.
The subjective evaluation uses pairwise ranking focused on reverberation similarity, so a future test that also measures overall perceived quality would clarify whether reverb preservation comes at any cost to other perceptual dimensions.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Colleague, here is my read on ReverbMiipher (arXiv:2505.05077). The core idea is simple: add a ReverbEncoder to Miipher-2, condition the vocoder with a 512-dim reverb vector, and during training randomly replace that vector with zeros so the model occasionally has to output clean speech. The subjective ranking (2,878 ratings) and the MCD/SPK-sim numbers in Table 2 do show that ReverbMiipher preserves reverberation better than Miipher-RIR, a baseline that is actually given an oracle RIR. That is a credible result, and the architecture is a natural extension rather than a wild leap. Credit where due: the PCA visualization in Fig. 3 shows the feature varies smoothly with RT60 and DRR, which supports the idea that the feature is at least reverb-related, and the paper is honest about the baseline being an upper bound on what real RIR estimation could achieve. My main concern is the disentanglement claim. Section 3.2 says the stochastic zero-vector replacement 'ensures' the feature encodes only reverberation, but Eq. (1) only trains the vocoder to interpret a zero vector as 'no reverb'. It imposes no constraint on what nonzero c can carry. The paper's own wording in §3.2 is hedged ('expected to be updated'), which is accurate, because there is no gradient pressure preventing c from leaking speaker identity, content, or residual noise. The objective metrics compare against s*r, so leakage that actually helps reconstruct the reverberant target would not be penalized. The controllability demonstrations in Section 5 are qualitative spectrograms; nothing there shows that interpolation changes only the reverb and not the speaker or wording. If c leaks, the 'novel reverberation' demos could be partially changing other attributes while appearing to control only reverb. This is not fatal to the preservation claim, but it does undermine the headline controllability claim as stated. A second, smaller gap: the paper says ReverbMiipher 'removes other artifacts' but never reports standard enhancement metrics like PESQ or STOI. The evidence is indirect (subjective resemblance to s*r). Given the architecture, the artifact removal is plausible, but it is not measured. Reproducibility is also partial: USM, WaveFit, the internal noise dataset, and the trained model are not released, though demo audio is. For a speech-restoration audience, this is worth engaging with. The reverb-preservation result is a genuine addition, and the training trick is worth discussing even if the disentanglement needs sharper verification. I would send it to peer review and ask the authors to do two things: include feature-leakage checks (e.g., show that speaker/content metrics are unchanged across c interpolation or swapping), and add at least one conventional speech quality metric like PESQ to back the artifact-removal claim. If they can supply a disentanglement analysis, this could become a solid conditional accept. If not, it remains a useful empirical contribution with an over-claimed controllability narrative. Recommended for a serious referee.

Referee Report

3 major / 5 minor

Summary. The paper proposes ReverbMiipher, an extension of the Miipher-2 speech restoration model that adds a ReverbEncoder producing a 512-dimensional utterance-level feature c from the noisy log-mel input. The vocoder is conditioned on c, and training uses stochastic zero-vector replacement so that setting c to the zero vector yields clean speech while using the extracted c yields speech with reverberation. Experiments compare ReverbMiipher against Miipher-2 and a two-stage baseline (Miipher-2 followed by convolution with an oracle-parameter simulated RIR) using pairwise subjective ranking and objective MCD/GPE/SPK-sim against the reverberant reference. The paper also presents PCA visualization, linear interpolation of c, and sampling from the PCA plane as reverberation control applications.

Significance. If the claimed disentanglement holds, ReverbMiipher is a meaningful step toward speech restoration that preserves and controls the acoustic environment rather than removing it. The strongest evidence is the subjective ranking (mean rank 1.49 versus 2.01 and 2.50, p < 1e-8) and the improvement in MCD and SPK-sim over an upper-bound two-stage baseline that is given oracle RT60 and DRR values. The proposed use of a learned reverb feature with interpolation and PCA sampling is a practical and potentially generalizable interface. However, the paper's central controllability claim rests on an untested disentanglement assumption, and the artifact-removal claim lacks direct objective metrics.

major comments (3)

[§3.2, Eq. (1)] The stochastic zero-vector replacement does not enforce the claimed disentanglement. In the branch U(0,1) <= q, the loss L_V(s, V(h_s, 0)) updates only the vocoder, not the ReverbEncoder, since c is fixed to 0. In the other branch, the ReverbEncoder is trained solely to minimize L_V(s*r, V(h_s, c)); any information in c that helps reconstruct the reverberant target, including speaker identity, content, or residual noise, is useful for this loss. The statement that artifacts in c 'will be treated as noise by the network' (§3.2) is an assertion without gradient-level justification. Since h_s is an imperfect estimate, c has a strong incentive to carry content and speaker information as a shortcut. Consequently, the interpolation and sampling demonstrations in §5 may change speaker or content rather than reverberation only. The authors should provide a quantitative disentanglement test, e.g., swap or interpolate c between utterances with different speakers or content and measure speaker similarity and word error rate of the output, or add an explicit information-bottleneck or adversarial objective.
[§4.4, Table 2] The abstract and §1 claim that ReverbMiipher 'removes other artifacts' such as noise and codec distortions, but Table 2 reports only MCD, GPE, and SPK-sim computed against the reverberant reference s*r. These metrics do not directly measure artifact removal (e.g., noise audibility, codec distortion) and are also consistent with c leaking content or speaker information, since such leakage would help match s*r. I recommend adding standard restoration metrics such as PESQ, STOI, or DNSMOS against the clean reference (or a no-reference metric), and possibly a subjective quality MOS, so that the artifact-removal claim is supported independently of the reverberation-preservation claim.
[§5.2–§5.3, Figs. 4–5] The controllability experiments are qualitative only. Fig. 4b and Fig. 5 show spectrograms and state that harmonic structure or original speech structure is preserved, but no objective measurement of content or speaker consistency is reported. Given the disentanglement concern in §3.2, these demonstrations do not establish that varying alpha or sampling c controls only reverberation. I suggest measuring, across the interpolation path alpha in [0,1] and across sampled points, word error rate (or phoneme error rate) and speaker embedding cosine similarity against the source utterance, as well as an objective reverb-parameter estimate (e.g., RT60 and DRR) of the output.

minor comments (5)

[§4.2, §6] There are minor typos: 'V orbis' should be 'Vorbis' in §4.2, and 'better than to the baseline' should be 'better than the baseline' in §6.
[Table 2] The objective results are reported without error bars or confidence intervals; with 1070 evaluation utterances, pairwise significance tests would strengthen the objective comparison.
[Fig. 3] Clarify the treatment of anechoic samples, since RT60 is nominally 0 and DRR is undefined for an anechoic condition; the text says anechoic samples cluster at upper right, which is confusing given the axis ranges.
[§5.2] Typo: 'uttrances' should be 'utterances', and 'Let c1 and c2 be a reverb features' should be 'Let c1 and c2 be reverb features'.
[§3.2] The switching probability q is not ablated; a brief sensitivity study or a justification for q=0.1 would help, since q controls the tradeoff between preservation and controllability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: ReverbMiipher's central claims rest on external supervised targets and held-out benchmarks.

full rationale

The paper's core derivation chain is supervised learning against external ground-truth signals, not a self-referential prediction. Equation (1) trains ReverbMiipher with targets s*r and s, both defined independently of the model, and the objective evaluations in Table 2 compare model outputs to the held-out reverberant reference s*r; the subjective evaluation in Table 1 compares listener rankings against the same external reference. The reverb feature c is a learned representation, not a fitted parameter renamed as a prediction; the interpolation and PCA-sampling demonstrations in Section 5 are empirical case studies rather than derived predictions. The paper cites prior work by overlapping authors ([26] Miipher-2, [45] WaveFit, [48] FLEURS-R) as building blocks, but these are pre-existing components with their own external training objectives and are not invoked as unverified uniqueness theorems or ansatz-smuggling citations. The disentanglement of c is asserted via the stochastic zero-vector strategy rather than rigorously demonstrated, which is a validity or robustness concern, but it is not a circular reduction: no equation in the paper is equivalent to its own input by construction, and no fitted parameter is passed off as a prediction. Therefore the appropriate circularity score is 0.

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

The reverb-feature vector is a learned latent representation, not a postulated physical entity. No new particles, forces, or conserved quantities are introduced. The free parameters are architecture and training hyperparameters chosen by hand.

free parameters (3)

switching probability q = 0.1
Chosen by hand; controls how often the model is trained to output anechoic speech with a zero reverb vector. The paper does not report tuning or sensitivity of this value.
reverb feature dimension D = 512
Architecture choice; no justification is provided for this dimensionality.
number of ReverbEncoder Conformer layers = 4
Architecture choice; no ablation study is reported.

assumptions (4)

domain assumption The clean SSL feature hs predicted by the feature cleaner is anechoic and contains all information needed to reconstruct the clean speech s.
Section 3.2 states that the training assumes hs exclusively contains speech information devoid of reverberation, noise, or other artifacts. ReverbMiipher inherits this from Miipher-2.
domain assumption A single 512-dimensional time-pooled vector can encode the relevant reverberation characteristics of an RIR from a noisy log-mel spectrogram.
This is central to the method, but the paper provides no analysis of the capacity or invariance of this representation.
ad hoc to paper Gradient updates with stochastic zero-vector replacement converge to a solution where c carries only reverberation and the vocoder ignores any leakage.
This is the proposed mechanism, but no quantitative disentanglement metric is given in Section 3.2 or Section 5.
domain assumption Image-method-simulated RIRs used in training are representative enough of real RIRs from the MIT survey at test time.
Training uses pyroomacoustics RIRs, evaluation uses real RIRs, and the domain shift is not analyzed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of ReverbMiipher: Generative Speech Restoration meets Reverberation Characteristics Controllability." pith.science (2026). https://pith.science/paper/BQO7R24T

@misc{pith2026250505077,
  author       = {Pith},
  title        = {Pith review of: ReverbMiipher: Generative Speech Restoration meets Reverberation Characteristics Controllability},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BQO7R24T}},
  note         = {Machine review of arXiv:2505.05077}
}

read the original abstract

Reverberation encodes spatial information regarding the acoustic source environment, yet traditional Speech Restoration (SR) usually completely removes reverberation. We propose ReverbMiipher, an SR model extending parametric resynthesis framework, designed to denoise speech while preserving and enabling control over reverberation. ReverbMiipher incorporates a dedicated ReverbEncoder to extract a reverb feature vector from noisy input. This feature conditions a vocoder to reconstruct the speech signal, removing noise while retaining the original reverberation characteristics. A stochastic zero-vector replacement strategy during training ensures the feature specifically encodes reverberation, disentangling it from other speech attributes. This learned representation facilitates reverberation control via techniques such as interpolation between features, replacement with features from other utterances, or sampling from a latent space. Objective and subjective evaluations confirm ReverbMiipher effectively preserves reverberation, removes other artifacts, and outperforms the conventional two-stage SR and convolving simulated room impulse response approach. We further demonstrate its ability to generate novel reverberation effects through feature manipulation.

Figures

Figures reproduced from arXiv: 2505.05077 by the authors.

**Figure 3.** PCA-based latent space distribution of reverb feature with [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

**Figure 5.** Reverb feature sampling from PCA-based latent space. The [PITH_FULL_IMAGE:figures/full_fig_p004_5.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

59 extracted references · 54 canonical work pages

[1]

Kuttruff, Room Acoustics

H. Kuttruff, Room Acoustics . Wiley, 1973

work page 1973
[2]

The effect of reverberation and audio spatialization on egocentric distance estimation of objects in stereoscopic virtual reality,

W. Bailey and B. M. Fazenda, “The effect of reverberation and audio spatialization on egocentric distance estimation of objects in stereoscopic virtual reality,” J. Acoust. Soc. Am. , 2017

work page 2017
[3]

Direct comparison of the impact of head tracking, reverberation, and individualized head- related transfer functions on the spatial perception of a virtual speech source,

D. R. Begault, E. M. Wenzel, and M. R. Anderson, “Direct comparison of the impact of head tracking, reverberation, and individualized head- related transfer functions on the spatial perception of a virtual speech source,” J. AES, 2001

work page 2001
[4]

Auditory-induced presence in mixed reality environments and related technology,

P. Larsson, A. V ¨aljam¨ae, et al. , “Auditory-induced presence in mixed reality environments and related technology,” The engineering of mixed reality systems, 2010

work page 2010
[5]

High-resolution image synthesis with latent diffusion models,

R. Rombach, A. Blattmann, et al., “High-resolution image synthesis with latent diffusion models,” in CVPR, 2022

work page 2022
[6]

AudioLM: a language modeling approach to audio generation,

Z. Borsos, R. Marinier, et al., “AudioLM: a language modeling approach to audio generation,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2023

work page 2023
[7]

P. A. Naylor and N. D. Gaubitch, Speech Dereverberation. Springer, 2010

work page 2010
[8]

Speech dereverberation based on variance-normalized delayed linear prediction,

T. Nakatani, T. Yoshioka, et al. , “Speech dereverberation based on variance-normalized delayed linear prediction,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2010

work page 2010

Show all 59 references

[9]

The REVERB challenge: A common evaluation framework for dereverberation and recognition of reverberant speech,

K. Kinoshita, M. Delcroix, et al. , “The REVERB challenge: A common evaluation framework for dereverberation and recognition of reverberant speech,” in WASPAA, 2013

2013
[10]

Learning spectral mapping for speech dereverberation,

K. Han, Y . Wang, and D. Wang, “Learning spectral mapping for speech dereverberation,” in ICASSP, 2014

2014
[11]

Neural network-based spectrum estimation for online wpe dereverberation

K. Kinoshita, M. Delcroix, et al. , “Neural network-based spectrum estimation for online wpe dereverberation.” in Interspeech, 2017

2017
[12]

Fundamentals, present and future perspectives of speech enhancement,

N. Das, S. Chakraborty, et al. , “Fundamentals, present and future perspectives of speech enhancement,” Int. J. of Speech Tech. , 2021

2021
[13]

Parametric resynthesis with neural vocoders,

S. Maiti and M. I. Mandel, “Parametric resynthesis with neural vocoders,” in WASPAA, 2019

2019
[14]

SelfRemaster: Self-supervised speech restoration with analysis-by-synthesis approach using channel modeling,

T. Saeki, S. Takamichi, et al. , “SelfRemaster: Self-supervised speech restoration with analysis-by-synthesis approach using channel modeling,” in Interspeech, 2022

2022
[15]

HiFi-GAN-2: Studio-quality speech enhancement via generative adversarial networks conditioned on acoustic features,

J. Su, Z. Jin, and A. Finkelstein, “HiFi-GAN-2: Studio-quality speech enhancement via generative adversarial networks conditioned on acoustic features,” in WASPAA, 2021

2021
[16]

V oiceFixer: A unified framework for high-fidelity speech restoration,

H. Liu, X. Liu, et al. , “V oiceFixer: A unified framework for high-fidelity speech restoration,” in Interspeech, 2022

2022
[17]

Universal speech enhancement with score- based diffusion,

J. Serr `a, S. Pascual, et al. , “Universal speech enhancement with score- based diffusion,” arXiv preprint arXiv:2206.03065 , 2022

2022 arXiv
[18]

Miipher: A robust speech restoration model integrating self-supervised speech and text representations,

Y . Koizumi, H. Zen, et al. , “Miipher: A robust speech restoration model integrating self-supervised speech and text representations,” in WASPAA, 2023

2023
[19]

Speech enhancement and dereverberation with diffusion-based generative models,

J. Richter, S. Welker, et al. , “Speech enhancement and dereverberation with diffusion-based generative models,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2023

2023
[20]

Diffusion models for audio restoration: A review,

J.-M. Lemercier, J. Richter, et al., “Diffusion models for audio restoration: A review,” IEEE Signal Process Mag. , 2024

2024
[21]

Universal score-based speech enhancement with high content preservation,

R. Scheibler, Y . Fujita,et al., “Universal score-based speech enhancement with high content preservation,” in Interspeech, 2024

2024
[22]

LLaSE-G1: Incentivizing generaliza- tion capability for llama-based speech enhancement,

B. Kang, X. Zhu, et al. , “LLaSE-G1: Incentivizing generaliza- tion capability for llama-based speech enhancement,” arXiv preprint arXiv:2503.00493, 2025

2025 arXiv
[23]

Genhancer: High-fidelity speech enhancement via generative modeling on discrete codec tokens,

H. Yang, J. Su, et al. , “Genhancer: High-fidelity speech enhancement via generative modeling on discrete codec tokens,” in Interspeech, 2024

2024
[24]

Joint semantic knowledge distillation and masked acoustic modeling for full-band speech restoration with improved intelligibility,

X. Liu, X. Li, et al. , “Joint semantic knowledge distillation and masked acoustic modeling for full-band speech restoration with improved intelligibility,” in ICASSP, 2025

2025
[25]

DiTSE: High-fidelity generative speech enhancement via latent diffusion transformers,

H. R. Guimar ˜aes, J. Su, et al. , “DiTSE: High-fidelity generative speech enhancement via latent diffusion transformers,” arXiv preprint arXiv:2504.09381, 2025

2025
[26]

Miipher-2: A universal speech restora- tion model for million-hour scale data restoration,

S. Karita, Y . Koizumi, et al. , “Miipher-2: A universal speech restora- tion model for million-hour scale data restoration,” arXiv preprint arXiv:2505.04457, 2025

2025 arXiv
[27]

Photorealistic video generation with diffusion models,

A. Gupta, L. Yu, et al. , “Photorealistic video generation with diffusion models,” in ECCV, 2023

2023
[28]

Stable video diffusion: Scal- ing latent video diffusion models to large datasets,

A. Blattmann, T. Dockhorn, et al. , “Stable video diffusion: Scal- ing latent video diffusion models to large datasets,” arXiv preprint arXiv:2311.15127, 2023

2023 arXiv
[29]

Image method for efficiently simulating small-room acoustics,

J. B. Allen and D. A. Berkley, “Image method for efficiently simulating small-room acoustics,” J. Acoust. Soc. Am. , 1979

1979
[30]

Simulating room transfer functions between transducers mounted on audio devices using a modified image source method,

Z. Xu, A. Herzog, et al. , “Simulating room transfer functions between transducers mounted on audio devices using a modified image source method,” The J. of the Acoust. Soc. Am. , 2024

2024
[31]

Fdtd methods for 3-d room acoustics simulation with high-order accuracy in space and time,

B. Hamilton and S. Bilbao, “Fdtd methods for 3-d room acoustics simulation with high-order accuracy in space and time,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2017

2017
[32]

Learning neural acoustic fields,

A. Luo, Y . Du, et al. , “Learning neural acoustic fields,” in NeurIPS, 2022

2022
[33]

IR-GAN: Room Impulse Response Generator for Far-Field Speech Recognition,

A. Ratnarajah, Z. Tang, and D. Manocha, “IR-GAN: Room Impulse Response Generator for Far-Field Speech Recognition,” in Interspeech, 2021

2021
[34]

Filtered noise shaping for time domain room impulse response estimation from reverberant speech,

C. J. Steinmetz, V . K. Ithapu, and P. Calamia, “Filtered noise shaping for time domain room impulse response estimation from reverberant speech,” in WASPAA, 2021

2021
[35]

Reverberation-controllable voice conver- sion using reverberation time estimator,

Y . Choi, C. Xie, and T. Toda, “Reverberation-controllable voice conver- sion using reverberation time estimator,” in Interspeech, 2023

2023
[36]

Multi-source spatial knowledge understanding for immersive visual text-to-speech,

S. He and R. Liu, “Multi-source spatial knowledge understanding for immersive visual text-to-speech,” in ICASSP, 2025

2025
[37]

Supervised speech separation based on deep learning: An overview,

D. Wang and J. Chen, “Supervised speech separation based on deep learning: An overview,” IEEE/ACM Trans. Audio, Speech and Lang. Proc., 2018

2018
[38]

A survey and an extensive evaluation of popular audio declipping methods,

P. Z´aviˇska, P. Rajmic, et al. , “A survey and an extensive evaluation of popular audio declipping methods,” IEEE Journal of Selected Topics in Signal Processing, 2020

2020
[39]

A deep neural network approach to speech bandwidth expansion,

K. Li and C.-H. Lee, “A deep neural network approach to speech bandwidth expansion,” in ICASSP, 2015

2015
[40]

Audio super-resolution using neural nets,

V . Kuleshov, S. Z. Enam, and S. Ermon, “Audio super-resolution using neural nets,” in ICLR (Workshop Track), 2017

2017
[41]

Neurips 2024 competition proposal: Urgent challenge,

W. Zhang, R. Scheibler, et al. , “Neurips 2024 competition proposal: Urgent challenge,” in NeurIPS 2024 Competition Track

2024
[42]

Google USM: Scaling automatic speech recognition beyond 100 languages,

Y . Zhang, W. Han, et al. , “Google USM: Scaling automatic speech recognition beyond 100 languages,” 2023

2023
[43]

Towards a unified view of parameter-efficient transfer learning,

J. He, C. Zhou, et al. , “Towards a unified view of parameter-efficient transfer learning,” in ICLR, 2022

2022
[44]

Conformer: Convolution-augmented transformer for speech recognition,

A. Gulati, C.-C. Chiu, et al. , “Conformer: Convolution-augmented transformer for speech recognition,” in Interspeech, 2020

2020
[45]

WaveFit: An iterative and non- autoregressive neural vocoder based on fixed-point iteration,

Y . Koizumi, K. Yatabe, et al. , “WaveFit: An iterative and non- autoregressive neural vocoder based on fixed-point iteration,” in SLT, 2023

2023
[46]

Pyroomacoustics: A python package for audio room simulation and array processing algorithms,

R. Scheibler, E. Bezzam, and I. Dokmani ´c, “Pyroomacoustics: A python package for audio room simulation and array processing algorithms,” in ICASSP, 2018

2018
[47]

Single-channel blind direct-to- reverberation ratio estimation using masking

W. Mack, S. Deng, and E. A. Habets, “Single-channel blind direct-to- reverberation ratio estimation using masking.” in Interspeech, 2020

2020
[48]

FLEURS-R: A restored multilingual speech corpus for generation tasks,

M. Ma, Y . Koizumi, et al. , “FLEURS-R: A restored multilingual speech corpus for generation tasks,” in Interspeech, 2024

2024
[49]

EARS: An anechoic fullband speech dataset benchmarked for speech enhancement and dereverberation,

J. Richter, Y .-C. Wu, et al. , “EARS: An anechoic fullband speech dataset benchmarked for speech enhancement and dereverberation,” in Interspeech, 2024

2024
[50]

WHAM!: Extending speech separation to noisy environments,

G. Wichern, J. Antognini, et al. , “WHAM!: Extending speech separation to noisy environments,” in Interspeech, 2019

2019
[51]

Statistics of natural reverberation enable perceptual separation of sound and space,

J. Traer and J. H. McDermott, “Statistics of natural reverberation enable perceptual separation of sound and space,” Proceedings of the National Academy of Sciences , 2016

2016
[52]

Adam: A method for stochastic optimization

D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization.” in ICLR, 2015

2015
[53]

W. J. Conover, Practical nonparametric statistics . john wiley & sons, 1999

1999
[54]

Multiple comparisons among means,

O. J. Dunn, “Multiple comparisons among means,” Journal of the American statistical association , 1961

1961
[55]

Mel-cepstral distance measure for objective speech quality assessment,

R. Kubichek, “Mel-cepstral distance measure for objective speech quality assessment,” in PACRIM, 1993

1993
[56]

Reducing f0 frame error of f0 tracking algorithms under noisy conditions with an unvoiced/voiced classification frontend,

W. Chu and A. Alwan, “Reducing f0 frame error of f0 tracking algorithms under noisy conditions with an unvoiced/voiced classification frontend,” in ICASSP, 2009

2009
[57]

Transfer learning from speaker verification to multispeaker text-to-speech synthesis,

Y . Jia, Y . Zhang,et al. , “Transfer learning from speaker verification to multispeaker text-to-speech synthesis,” in NeurIPS, 2018

2018
[58]

Sample efficient adaptive text-to-speech,

Y . Chen, Y . Assael,et al. , “Sample efficient adaptive text-to-speech,” in ICLR, 2019

2019
[59]

New method of measuring reverberation time,

M. R. Schroeder, “New method of measuring reverberation time,” The Journal of the Acoustical Society of America , 1965

1965

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Kuttruff, Room Acoustics

H. Kuttruff, Room Acoustics . Wiley, 1973

work page 1973

[2] [2]

The effect of reverberation and audio spatialization on egocentric distance estimation of objects in stereoscopic virtual reality,

W. Bailey and B. M. Fazenda, “The effect of reverberation and audio spatialization on egocentric distance estimation of objects in stereoscopic virtual reality,” J. Acoust. Soc. Am. , 2017

work page 2017

[3] [3]

Direct comparison of the impact of head tracking, reverberation, and individualized head- related transfer functions on the spatial perception of a virtual speech source,

D. R. Begault, E. M. Wenzel, and M. R. Anderson, “Direct comparison of the impact of head tracking, reverberation, and individualized head- related transfer functions on the spatial perception of a virtual speech source,” J. AES, 2001

work page 2001

[4] [4]

Auditory-induced presence in mixed reality environments and related technology,

P. Larsson, A. V ¨aljam¨ae, et al. , “Auditory-induced presence in mixed reality environments and related technology,” The engineering of mixed reality systems, 2010

work page 2010

[5] [5]

High-resolution image synthesis with latent diffusion models,

R. Rombach, A. Blattmann, et al., “High-resolution image synthesis with latent diffusion models,” in CVPR, 2022

work page 2022

[6] [6]

AudioLM: a language modeling approach to audio generation,

Z. Borsos, R. Marinier, et al., “AudioLM: a language modeling approach to audio generation,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2023

work page 2023

[7] [7]

P. A. Naylor and N. D. Gaubitch, Speech Dereverberation. Springer, 2010

work page 2010

[8] [8]

Speech dereverberation based on variance-normalized delayed linear prediction,

T. Nakatani, T. Yoshioka, et al. , “Speech dereverberation based on variance-normalized delayed linear prediction,” IEEE/ACM Trans. Audio, Speech and Lang. Proc. , 2010

work page 2010