Pith. sign in

REVIEW 4 major objections 4 minor 36 references

Evaluating the Impact of Discriminative and Generative E2E Speech Enhancement Models on Syllable Stress Preservation

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Generative speech enhancement preserves syllable stress better than discriminative models.

desk verdict Useful first comparison of SE models for stress preservation, but the central CDiffuSE ranking is unsupported by unreplicated point estimates and contradicted by Table II at 0 dB. read the letter →

arxiv 2412.08306 v1 pith:SQOZQH2R submitted 2024-12-11 eess.AS cs.SD

Rangavajjala Sankara Bharadwaj , Jhansi Mallela , Sai Harshitha Aluru , Chiranjeevi Yarra This is my paper · ORCID
classification eess.AScs.SD
keywords syllablestressdetectionspeechenhancementdiffusionmodelsdiscriminativevsgenerativeheuristicacousticfeatureswav2vec2.0ISLEcorpusperceptualstudy
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

The paper asks whether speech enhancement, used to clean noisy audio before automatic syllable stress detection, preserves the stress patterns that listeners and language-learning systems depend on. It compares three enhancement models—two discriminative and one generative—on English speech from non-native German and Italian speakers at SNRs from 0 to 20 dB. The central claim is that the generative diffusion-based model (CDiffuSE) keeps stress detection accuracy highest, that heuristic acoustic and context features stay robust under noise while wav2vec 2.0 features degrade sharply at low SNR, and that human listeners rank the same model best. If this holds, generative enhancement is the better front-end for stress-focused CALL systems, and heuristic features are safer than self-supervised ones in noisy conditions.

What carries the argument

The pipeline computes syllable boundaries once on clean audio using linguist-corrected phoneme alignments and a syllabification tool, then applies those timestamps to noisy and enhanced audio to cut syllable segments. For each syllable it extracts either a 38-dimensional heuristic feature set (sonority-based prominence contours plus binary context features) or 768-dimensional wav2vec 2.0 frame averages, feeds them to a jointly optimized VAE+DNN classifier, and applies a postprocessing rule that keeps exactly one stressed syllable per word. The three speech enhancement models—CDiffuSE (generative diffusion), Denoiser (waveform-domain discriminative), and DTLN (real-time discriminative)—are the interventions being compared inside this fixed pipeline.

What would settle it

Re-run stress detection after re-aligning syllable boundaries on each enhanced waveform and compare model rankings; if CDiffuSE's advantage over Denoiser and DTLN shrinks or disappears, the headline ranking comes from clean-audio alignment timings rather than from genuine stress preservation.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is a ranking: the generative CDiffuSE model outperforms the discriminative DTLN and Denoiser models for preserving syllable stress across almost all tested SNR levels, and this ranking matches human perception. The authors also find that heuristic acoustic and context features are robust to noise and enhancement, while wav2vec 2.0 representations suffer large accuracy drops at 0 and 5 dB SNR, both on noisy and enhanced audio. The perceptual study reinforces the automatic results: listeners judged CDiffuSE-enhanced audio as most similar to clean speech, followed by Denoiser and then DTLN, and the same order appears in automatic stress detection accuracy on the subset used for listening.

Load-bearing premise

The load-bearing premise is that syllable boundaries and stress labels derived from clean audio remain correct after noise addition and enhancement, so the model never re-aligns to the processed signal and any timing shift would count as a stress-preservation failure instead of an alignment artifact.

Editorial extensions

If this is right

  • CALL systems operating in noisy environments can use CDiffuSE as a preprocessing front-end to keep syllable stress detection accuracy close to clean-speech levels.
  • Heuristic sonority-based features are the safer choice at low SNRs, contrary to the common expectation that self-supervised representations like wav2vec 2.0 dominate traditional features.
  • Human perception tracks automatic stress detection, so automatic accuracy on enhanced speech is a reliable proxy for what learners hear.
  • Discriminative enhancers can actively hurt stress detection compared to leaving speech noisy, especially on the Italian dataset, meaning enhancement is not always beneficial for this task.
  • The one-stressed-syllable postprocessing rule becomes more important when wav2vec 2.0 features are used at 0 dB, where raw classification is least reliable.

Reading between the lines

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

  • Because syllable boundaries come from clean audio and are never re-estimated on enhanced signals, the reported ranking could partly reflect how well each enhancer preserves timing rather than stress cues; a boundary re-alignment test would separate these effects.
  • wav2vec 2.0 features are averaged over entire syllables, so additive Gaussian noise is averaged into every feature dimension, which may explain their severe low-SNR degradation; frame-level classifiers or noise-augmented training could change that comparison.
  • CDiffuSE's iterative denoising may act as a prosody-preserving prior rather than just a noise remover; a direct extension would use its learned representations for stress detection without explicit syllable segmentation.
  • The perceptual study used only 50 words and 25 listeners, so the alignment between human and automatic judgment is promising but would need a larger stimulus set to generalize across speakers and word types.
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.

Referee Report

4 major / 4 minor

Summary. The paper evaluates three end-to-end speech enhancement (SE) models—DTLN and Denoiser (discriminative) and CDiffuSE (generative)—as preprocessors for automatic syllable stress detection on the ISLE corpus (German and Italian learners of English). Noisy speech at 0, 5, 10, and 20 dB SNR is enhanced by each model, and a jointly optimized VAE+DNN classifier is trained on either 38-dimensional heuristic acoustic/context features or 768-dimensional wav2vec 2.0 features, with and without a one-stressed-syllable postprocessing step. A perceptual study with 25 listeners compares how similar enhanced audio is to clean audio in terms of stress placement. The central claims are that (1) heuristic features are more noise-robust than wav2vec 2.0 features, (2) the generative model CDiffuSE outperforms the discriminative models for stress detection, and (3) perceptual results align with automatic detection results.

Significance. If the central ranking were statistically well supported, the paper would provide a useful practical recommendation: use CDiffuSE as an SE front-end for syllable stress detection in CALL systems and prefer heuristic features over wav2vec 2.0 under low SNR. The use of a real L2 corpus (ISLE), two language groups, three SE models, two feature families, and a listener study are strengths, as is the focus on a relatively understudied downstream task (prosodic stress preservation rather than ASR or intelligibility). However, the paper's main conclusion rests on point estimates from a single 5-fold run with no confidence intervals or significance tests, and the perceptual component lacks inter-rater reliability. Because the reported margins are often below 0.5 percentage points and one feature condition reverses the ranking, the central claims are not yet established to the standard needed for a journal publication.

major comments (4)
  1. [§V-A, Table I] The central claim that CDiffuSE outperforms the discriminative models is supported only by point estimates from a single 5-fold cross-validation run, with no standard deviations, confidence intervals, or paired significance tests. Several GER differences are very small: at 0 dB CDiffuSE beats Denoiser by 0.45 pp, at 5 dB by 0.15 pp, at 10 dB by 0.00 pp, and at 20 dB by 0.15 pp (without postprocessing); with postprocessing the gaps are 0.5, 0.25, 0.4, and 0.55 pp. These margins are well within the run-to-run variability expected for a VAE+DNN trained on roughly 3.7k utterances. The paper must report variability across seeds or folds and a paired test (e.g., McNemar or bootstrap over utterances) before the ranking can be accepted.
  2. [§V-A, Table II and §VI] The conclusion that 'audios enhanced with the SE model belonging to generative modeling category (CDiffuSE) outperforms the discriminative SE models' is stated without qualification, but Table II shows that with wav2vec 2.0 features at 0 dB, Denoiser beats CDiffuSE in both languages (GER: 91.42 vs 89.40; ITA: 91.40 vs 90.30). The paper's own abstract limits the robustness claim to 'when heuristic features are used,' but the conclusion does not. The conclusion should be restricted to the heuristic-feature condition or should provide an explanation for the reversal under self-supervised features.
  3. [§V-B, Table III] The perceptual study is presented as confirming the automatic results, but no inter-rater reliability, no confidence intervals, and no significance test are reported for the listener preference percentages (45.91% vs 30.34% vs 23.75% for ITA; 37.60% vs 34.88% vs 27.52% for GER). With only 25 items per language and one vote per item per listener, the margins between Denoiser and DTLN in GER (34.88% vs 27.52%) may not be meaningful. In addition, the automatic accuracies in Table III (e.g., ITA CDiffuSE 82.65%) differ substantially from the corresponding Table I values (around 92.4–93.0%), so the perceptual subset is not representative of the main evaluation and the alignment between the two tables is not explained.
  4. [§II and §III-A] The evaluation assumes that syllable boundaries computed on clean speech remain valid for noisy and enhanced audio. The dataset section states that phoneme alignments are computed on clean recordings and corrected by linguists, and Section III-A says syllable segmentation is performed 'on each speech signal using syllable timestamps' without re-alignment. If enhancement shifts timing or introduces artifacts, the features are computed on misaligned segments, and accuracy differences may reflect alignment error rather than SE quality. The paper should either re-align enhanced audio or provide evidence that the clean timestamps remain accurate after enhancement.
minor comments (4)
  1. [§V-A] The cross-references in the text appear as 'Table ??' and 'Table ??' in both subsections of Section V-A; these placeholders should be replaced with the actual table numbers.
  2. [Throughout] There are several typographical errors, including 'discirminative' in the introduction, 'Wave2vec 2.0' in the Table II header, 'on on the V oiceBank-DEMAND dataset' in Section III-B3, and 'Adressing' in Section III-B. A careful proofreading pass is needed.
  3. [§III-C] The perceptual study section does not specify how the 50 word stimuli were chosen, how many times each subject listened to each sample, or whether the order of presentation was randomized; those details should be added for reproducibility.
  4. [§IV] The architecture description states that 'all the DNN and VAE parameters ... are optimal and we choose them by maximizing the performance on the validation set,' but no search grid or selection criterion is given; reporting the hyperparameter range would improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the central claim is an empirical comparison with no derivation chain that reduces to its inputs.

full rationale

The paper makes no formal derivation; its central claim -- that audios enhanced with the generative CDiffuSE model outperform discriminative DTLN and Denoiser for syllable stress detection -- is an empirical observation based on the accuracy tables. The VAE+DNN classifier from [12] and the wav2vec 2.0 stress features from [20] are used as fixed measuring instruments from prior work; they are not fitted within this paper and do not by construction determine the relative ranking of the three SE models. These self-citations identify the detector and feature extractor, but the manipulated conditions (noise type, SNR levels, choice of SE models) are external to those cited methods, so no load-bearing argument reduces to a self-citation chain. No equation or fitted parameter is renamed as a prediction, and no definition of the outcome is built into the inputs. The absence of confidence intervals or significance tests is a statistical robustness concern, not a circularity concern.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the measuring instruments: the clean-speech syllable alignments applied to enhanced audio, the VAE+DNN classifier as stress oracle, Gaussian noise as the only corruption, and a small perceptual panel. These are domain assumptions rather than fitted free parameters. The only fitted quantities are the undisclosed classifier hyperparameters selected on a validation split. No new entities are introduced.

free parameters (1)
  • VAE and DNN hyperparameters (number of layers, hidden units) = not reported, selected by maximizing validation performance
    Section IV states all DNN and VAE parameters are optimal and chosen by maximizing performance on the validation set, but the values and search procedure are not disclosed.
assumptions (4)
  • domain assumption Linguist-corrected phoneme and syllable alignments computed on clean speech remain valid for noisy and enhanced speech.
    Section II and Section III-A compute syllable segmentations on clean recordings and use those timestamps for every condition; no re-alignment or audit is performed after enhancement.
  • domain assumption The VAE+DNN stress detector from prior work [12] is a reliable proxy for stress preservation in enhanced audio.
    Section IV uses this jointly optimized VAE+DNN as the sole automatic evaluator, and clean-speech accuracy is taken as the reference for preservation.
  • domain assumption White Gaussian noise at 0 to 20 dB SNR is representative of real-world noise conditions for CALL systems.
    Section III-A generates noisy data using white Gaussian noise only; no babble, stationary, or other noise types are tested.
  • domain assumption A 25-subject perceptual similarity judgment on 50 selected words captures human stress perception.
    Section III-C relies on these ratings to claim alignment with automatic results, but no inter-rater reliability or statistical test is reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Evaluating the Impact of Discriminative and Generative E2E Speech Enhancement Models on Syllable Stress Preservation." pith.science (2026). https://pith.science/paper/SQOZQH2R

@misc{pith2026241208306,
  author       = {Pith},
  title        = {Pith review of: Evaluating the Impact of Discriminative and Generative E2E Speech Enhancement Models on Syllable Stress Preservation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SQOZQH2R}},
  note         = {Machine review of arXiv:2412.08306}
}
read the original abstract

Automatic syllable stress detection is a crucial component in Computer-Assisted Language Learning (CALL) systems for language learners. Current stress detection models are typically trained on clean speech, which may not be robust in real-world scenarios where background noise is prevalent. To address this, speech enhancement (SE) models, designed to enhance speech by removing noise, might be employed, but their impact on preserving syllable stress patterns is not well studied. This study examines how different SE models, representing discriminative and generative modeling approaches, affect syllable stress detection under noisy conditions. We assess these models by applying them to speech data with varying signal-to-noise ratios (SNRs) from 0 to 20 dB, and evaluating their effectiveness in maintaining stress patterns. Additionally, we explore different feature sets to determine which ones are most effective for capturing stress patterns amidst noise. To further understand the impact of SE models, a human-based perceptual study is conducted to compare the perceived stress patterns in SE-enhanced speech with those in clean speech, providing insights into how well these models preserve syllable stress as perceived by listeners. Experiments are performed on English speech data from non-native speakers of German and Italian. And the results reveal that the stress detection performance is robust with the generative SE models when heuristic features are used. Also, the observations from the perceptual study are consistent with the stress detection outcomes under all SE models.

Figures

Figures reproduced from arXiv: 2412.08306 by the authors.

Figure 1
Figure 1. Overview of the proposed approach for evaluating the [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Block diagram demonstrating automatic syllable stress detection and perceptual study conducted on SE-enhanced speech. The [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 29 canonical work pages

  1. [1]

    M Levy, Computer-assisted language learning: Context and conceptu- alization, Oxford University Press, 1997

    work page 1997

  2. [2]

    Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems,

    Luciana Ferrer, Harry Bratt, Colleen Richey, Horacio Franco, Victor Abrash, and Kristin Precoda, “Classification of lexical stress using spectral and prosodic features for computer-assisted language learning systems,” Speech Communication , vol. 69, pp. 31–45, 2015

    work page 2015

  3. [3]

    Jacob Benesty, Shoji Makino, and Jingdong Chen, Speech enhancement, Springer Science & Business Media, 2006

    work page 2006

  4. [4]

    Speech enhancement with lstm recurrent neural networks and its application to noise-robust ASR,

    Felix Weninger, Hakan Erdogan, Shinji Watanabe, Emmanuel Vincent, Jonathan Le Roux, John R Hershey, and Bj ¨orn Schuller, “Speech enhancement with lstm recurrent neural networks and its application to noise-robust ASR,” in Latent V ariable Analysis and Signal Separation: 12th International Conference, LVA/ICA 2015, Liberec, Czech Republic, August 25-28, 2...

    work page 2015

  5. [5]

    Towards robust speech emotion recognition using deep residual networks for speech enhancement,

    Andreas Triantafyllopoulos, Gil Keren, Johannes Wagner, Ingmar Steiner, and Bj¨orn Schuller, “Towards robust speech emotion recognition using deep residual networks for speech enhancement,” 2019

    work page 2019

  6. [6]

    Conditional Generative Adversarial Networks for Speech Enhancement and Noise-Robust Speaker Verification

    Daniel Michelsanti and Zheng-Hua Tan, “Conditional generative ad- versarial networks for speech enhancement and noise-robust speaker verification,” arXiv preprint arXiv:1709.01703 , 2017

    work page Pith review arXiv 2017

  7. [7]

    SLIM prosodic module for learning activities in a foreign language,

    Rodolfo Delmonte, Mirela Petrea, and Ciprian Bacalu, “SLIM prosodic module for learning activities in a foreign language,” in Fifth European Conference on Speech Communication and Technology , 1997

    work page 1997

  8. [8]

    Automatic syllable stress detection using prosodic features for pronunciation evaluation of language learners,

    Joseph Tepperman and Shrikanth Narayanan, “Automatic syllable stress detection using prosodic features for pronunciation evaluation of language learners,” in International Conference on Acoustics, Speech, and Signal Processing. IEEE, 2005, vol. 1, pp. 937–940

    work page 2005

Show all 36 references
  1. [9]

    Automatic prominent syllable detection with machine learning classifiers,

    David O Johnson and Okim Kang, “Automatic prominent syllable detection with machine learning classifiers,” International Journal of Speech Technology, 2015, vol. 18, no. 4, pp. 583–592

    2015

  2. [10]

    An Attention Based Deep Neural Network for Automatic Lexical Stress Detection,

    Tian Xia, Xianfeng Rui, Chien Lin Huang, Iek Heng Chu, Shaojun Wang, and Mei Han, “An Attention Based Deep Neural Network for Automatic Lexical Stress Detection,” in Global Conference on Signal and Information Processing (GlobalSIP). IEEE, 2019, pp. 1–5

    2019

  3. [11]

    An end-to-end approach for lexical stress detection based on transformer,

    Yong Ruan, Xiangdong Wang, Hong Liu, Zhigang Ou, Yun Gao, Jianfeng Cheng, and Yueliang Qian, “An end-to-end approach for lexical stress detection based on transformer,” arXiv preprint arXiv:1911.04862, 2019

    1911 arXiv

  4. [12]

    A compar- ison of learned representations with jointly optimized vae and dnn for syllable stress detection,

    Jhansi Mallela, Prasanth Sai Boyina, and Chiranjeevi Yarra, “A compar- ison of learned representations with jointly optimized vae and dnn for syllable stress detection,” in International Conference on Speech and Computer. Springer, 2023, pp. 322–334

    2023

  5. [13]

    The ISLE corpus of non-native spoken English,

    Wolfgang Menzel, Eric Atwell, Patrizia Bonaventura, Daniel Herron, Peter Howarth, Rachel Morton, and Clive Souter, “The ISLE corpus of non-native spoken English,” in Proceedings of LREC: Language Resources and Evaluation Conference, vol. 2. European Language Re- sources Associ...

    2000

  6. [14]

    Automatic detection of syllable stress using sonority based prominence features for pronunciation evaluation,

    Chiranjeevi Yarra, Om D Deshmukh, and Prasanta Kumar Ghosh, “Automatic detection of syllable stress using sonority based prominence features for pronunciation evaluation,” in International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2017, pp. 5845–5849

    2017

  7. [15]

    Comparison of automatic syllable stress detection quality with time-aligned boundaries and context dependencies,

    Chiranjeevi Yarra, Manoj Kumar Ramanathi, and Prasanta Kumar Ghosh, “Comparison of automatic syllable stress detection quality with time-aligned boundaries and context dependencies,” in SLaTE, 2019, pp. 79–83

    2019

  8. [16]

    wav2vec 2.0: A framework for self-supervised learning of speech representations,

    Alexei Baevski, Yuhao Zhou, Abdelrahman Mohamed, and Michael Auli, “wav2vec 2.0: A framework for self-supervised learning of speech representations,” Advances in neural information processing systems , vol. 33, pp. 12449–12460, 2020

    2020

  9. [17]

    Hubert: Self- supervised speech representation learning by masked prediction of hidden units,

    Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakho- tia, Ruslan Salakhutdinov, and Abdelrahman Mohamed, “Hubert: Self- supervised speech representation learning by masked prediction of hidden units,” Transactions on Audio, Speech, and Language Processing (ACM), v...

    2021

  10. [18]

    Multi-lingual multi-task speech emotion recognition using wav2vec 2.0,

    Mayank Sharma, “Multi-lingual multi-task speech emotion recognition using wav2vec 2.0,” in International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2022, pp. 6907–6911

    2022

  11. [19]

    A fine-tuned wav2vec 2.0/hubert benchmark for speech emotion recogni- tion, speaker verification and spoken language understanding,

    Yingzhi Wang, Abdelmoumene Boumadane, and Abdelwahab Heba, “A fine-tuned wav2vec 2.0/hubert benchmark for speech emotion recogni- tion, speaker verification and spoken language understanding,” arXiv preprint arXiv:2111.02735, 2021

    2021 arXiv

  12. [20]

    Exploring the use of self-supervised representations for automatic syllable stress detection,

    Jhansi Mallela, Sai Harshitha Aluru, and Chiranjeevi Yarra, “Exploring the use of self-supervised representations for automatic syllable stress detection,” in 2024 National Conference on Communications (NCC) . IEEE, 2024, pp. 1–6

    2024

  13. [21]

    Conv-tasnet: Surpassing ideal time– frequency magnitude masking for speech separation,

    Yi Luo and Nima Mesgarani, “Conv-tasnet: Surpassing ideal time– frequency magnitude masking for speech separation,” IEEE/ACM transactions on audio, speech, and language processing , vol. 27, no. 8, pp. 1256–1266, 2019

    2019

  14. [22]

    Real-time single-channel dereverbera- tion and separation with time-domain audio separation network.,

    Yi Luo and Nima Mesgarani, “Real-time single-channel dereverbera- tion and separation with time-domain audio separation network.,” in Interspeech, 2018, pp. 342–346

    2018

  15. [23]

    U-net: Con- volutional networks for biomedical image segmentation,

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox, “U-net: Con- volutional networks for biomedical image segmentation,” in Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, p...

    2015

  16. [24]

    Tasnet: time-domain audio separation network for real-time, single-channel speech separation,

    Yi Luo and Nima Mesgarani, “Tasnet: time-domain audio separation network for real-time, single-channel speech separation,” in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 2018, pp. 696–700

    2018

  17. [25]

    Conditional diffusion probabilistic model for speech enhancement,

    Yen-Ju Lu, Zhong-Qiu Wang, Shinji Watanabe, Alexander Richard, Cheng Yu, and Yu Tsao, “Conditional diffusion probabilistic model for speech enhancement,” in ICASSP 2022-2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2022, pp. 7402–7406

    2022

  18. [26]

    Dual-signal transforma- tion lstm network for real-time noise suppression,

    Nils L Westhausen and Bernd T Meyer, “Dual-signal transforma- tion lstm network for real-time noise suppression,” arXiv preprint arXiv:2005.07551, 2020

    2005 arXiv

  19. [27]

    Librispeech: an asr corpus based on public domain audio books,

    Vassil Panayotov, Guoguo Chen, Daniel Povey, and Sanjeev Khudanpur, “Librispeech: an asr corpus based on public domain audio books,” in 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, 2015, pp. 5206–5210

    2015

  20. [28]

    Audio set: An ontology and human-labeled dataset for audio events,

    Jort F Gemmeke, Daniel PW Ellis, Dylan Freedman, Aren Jansen, Wade Lawrence, R Channing Moore, Manoj Plakal, and Marvin Ritter, “Audio set: An ontology and human-labeled dataset for audio events,” in 2017 IEEE international conference on acoustics, speech and signal processing...

    2017

  21. [29]

    The di- verse environments multi-channel acoustic noise database (demand): A database of multichannel environmental noise recordings,

    Joachim Thiemann, Nobutaka Ito, and Emmanuel Vincent, “The di- verse environments multi-channel acoustic noise database (demand): A database of multichannel environmental noise recordings,” in Proceed- ings of Meetings on Acoustics . AIP Publishing, 2013, vol. 19

    2013

  22. [30]

    Music source separation in the waveform domain,

    Alexandre D ´efossez, Nicolas Usunier, L ´eon Bottou, and Francis Bach, “Music source separation in the waveform domain,” arXiv preprint arXiv:1911.13254, 2019

    1911 arXiv

  23. [31]

    Real time speech enhancement in the waveform domain,

    Alexandre Defossez, Gabriel Synnaeve, and Yossi Adi, “Real time speech enhancement in the waveform domain,” arXiv preprint arXiv:2006.12847, 2020

    2006 arXiv

  24. [32]

    Noisy speech database for training speech enhancement algorithms and tts models,

    Cassia Valentini-Botinhao et al., “Noisy speech database for training speech enhancement algorithms and tts models,” University of Edin- burgh. School of Informatics. Centre for Speech Technology Research (CSTR), 2017

    2017

  25. [33]

    The interspeech 2020 deep noise suppression challenge: Datasets, subjective testing framework, and challenge results,

    Chandan KA Reddy, Vishak Gopal, Ross Cutler, Ebrahim Beyrami, Roger Cheng, Harishchandra Dubey, Sergiy Matusevych, Robert Aich- ner, Ashkan Aazami, Sebastian Braun, et al., “The interspeech 2020 deep noise suppression challenge: Datasets, subjective testing framework, and chal...

    2020 arXiv

  26. [34]

    Deep unsupervised learning using nonequilibrium thermody- namics,

    Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Surya Ganguli, “Deep unsupervised learning using nonequilibrium thermody- namics,” in International conference on machine learning . PMLR, 2015, pp. 2256–2265

    2015

  27. [35]

    Investigating RNN-based speech enhancement methods for noise-robust text-to-speech.,

    Cassia Valentini-Botinhao, Xin Wang, Shinji Takaki, and Junichi Ya- magishi, “Investigating RNN-based speech enhancement methods for noise-robust text-to-speech.,” in SSW, 2016, pp. 146–152

    2016

  28. [36]

    The voice bank corpus: Design, collection and data analysis of a large regional accent speech database,

    Christophe Veaux, Junichi Yamagishi, and Simon King, “The voice bank corpus: Design, collection and data analysis of a large regional accent speech database,” in 2013 international conference oriental COCOSDA held jointly with 2013 conference on Asian spoken language research ...

    2013

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.