Enhancing Listened Speech Decoding from EEG via Parallel Phoneme Sequence Prediction

Aditya Kommineni; Jihwan Lee; Shrikanth Narayanan; Sudarsana Reddy Kadiri; Tiantian Feng

REVIEW 2 major objections 6 minor 28 references

Enhancing Listened Speech Decoding from EEG via Parallel Phoneme Sequence Prediction

T0 review · 2 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Adding an auxiliary phoneme-sequence predictor to an EEG-to-speech model lets one system decode listened speech as both audio and text in parallel, and improves both outputs.

desk verdict A promising multi-task EEG-to-speech/phone decoding result whose central attribution claim is not yet proven; needs a control ablation and a more precise abstract. read the letter →

arxiv 2501.04844 v1 pith:IONIODDO submitted 2025-01-08 eess.AS cs.AIcs.CLeess.SP

Jihwan Lee , Tiantian Feng , Aditya Kommineni , Sudarsana Reddy Kadiri , Shrikanth Narayanan This is my paper · ORCID

classification eess.AScs.AIcs.CLeess.SP

keywords EEGspeechdecodingphonemesequencepredictionparallelbrain-computerinterfacemulti-tasklearninglistenedCTClossaudioreconstruction

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 tries to show that a brain-computer interface can decode listened speech from EEG in two modalities at once—reconstructed audio waveforms and text-like phoneme sequences—and that asking the model to predict phonemes as a side task actually makes the audio reconstruction better. On the N400 dataset with held-out speakers and sentences, the proposed model roughly doubles the mel-spectrogram correlation of the decoded speech compared with the previous end-to-end system, while also beating a text-decoding baseline on phoneme top-1 accuracy in the harder test conditions. The authors interpret this as evidence that phoneme-level supervision sharpens the EEG embeddings in a way that benefits the primary speech reconstruction task. If true, the result matters for BCI communication because a single model could deliver both audible and textual output without chaining separate decoders.

What carries the argument

The load-bearing object is the auxiliary phoneme predictor: a conformer-block encoder followed by a single-layer LSTM with attention, fed the same EEG embeddings that the speech module uses, and trained with connectionist temporal classification (CTC) loss. Its total objective is $L = L_{EEG} + \alpha L_{CTC}$ with $\alpha = 0.3$, so the EEG module is pushed to preserve phoneme-discriminative structure in its embeddings. A gradient stop between the EEG module and the speech module means the speech branch does not influence the EEG encoder's training signal; the phoneme loss is what changes the shared embedding geometry relative to the previous system.

What would settle it

Train a control model identical to CB-0 but with the phoneme predictor fed randomly shuffled phoneme targets, or with a same-capacity branch trained only on EEG reconstruction; if mel-spectrogram correlation stays close to the reported 22–32% range, the gains are not caused by phoneme supervision and the central claim collapses. A matched-capacity speech-only baseline with no phoneme branch would also settle the attribution.

Watch

Extended reading notes

Core claim

The central claim is that parallel phoneme sequence prediction is not just an add-on output but a training signal that improves listened speech decoding from EEG. Concretely, the paper's configurations CB-0, CB-1, and CB-2, which share EEG and speech modules with the FESDE baseline but add a phoneme branch trained with CTC loss, reduce mel-cepstral distortion from about 11.7–11.8 dB to 9.6–11.3 dB and raise mel-spectrogram correlation from roughly 13–14% to 22–32% across the three held-out test sets. On phoneme decoding, the two-conformer-block configuration beats the adapted EEG2TEXT baseline on top-1 accuracy on all three test sets, with the largest margin on the unseen-both set. The paper also reports a trade-off: more conformer blocks improve phoneme decoding while slightly hurting speech reconstruction, yet even the strongest phoneme configuration keeps speech reconstruction above the previous baseline.

Load-bearing premise

The paper credits the phoneme-prediction task for the speech gains, but it never runs a control with the same extra capacity and no phoneme information, so the improvement could come from the extra parameters rather than the phoneme supervision.

Editorial extensions

If this is right

A single EEG-to-speech model can output both reconstructed audio and phoneme text simultaneously, removing the need to chain a speech synthesizer or recognizer after a single-modality decoder.
Phoneme-level supervision improves waveform reconstruction enough to roughly double mel-spectrogram correlation over the previous end-to-end baseline.
The choice of phoneme predictor capacity trades off the two modalities: one LSTM layer gives the best audio, while two conformer blocks give the best phoneme accuracy.
The phoneme-level analysis shows articulation classes differ in decodability, with nasal, dental, and voiced consonants easier in audio and front, close, and lax vowels easier in text.
The approach is currently limited to listened speech perception, and the authors point to imagined or attempted speech as the natural next step.

Reading between the lines

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

If the phoneme branch is genuinely doing the work, a natural testable extension is to use richer phoneme targets, such as those with stress or duration marks, or to add a language model over phonemes; the paper's own results suggest gains would concentrate on consonants, where the modality trade-off is largest.
The reported trade-off across conformer depth suggests that the auxiliary task's parameter count, not just the task itself, may be a confound; matched-capacity ablations with random or shuffled phoneme targets would separate phoneme supervision from extra parameters.
The high top-1 accuracy on the unseen-subject test set likely reflects sequence memorization, since the target phoneme sequences are already seen during training; future subject-independent evaluation should hold out sentences as well.
The same parallel architecture could be tried on imagined or attempted speech, where phoneme targets are less well time-aligned and CTC's many-to-one flexibility may behave differently.

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

Here's my take on 2501.04844. The headline: the authors take their own FESDE EEG-to-speech model, bolt on a CTC phoneme predictor that shares the EEG encoder, and report that speech reconstruction gets much better (Mel-Corr roughly doubles). That's a genuinely interesting result, and to my knowledge no one has shown parallel waveform-plus-phoneme decoding from EEG in one model. The code and samples are public, the writing is clear, and the gains on speech are consistent across all three test splits with confidence intervals. The phoneme-group analysis is a nice extra, and the stop-gradient between the EEG and speech modules is a sensible design choice.

The soft spots are real though. First, the central attribution isn't proven. The comparison is against FESDE, but the new model differs in two confounded ways: it adds an auxiliary branch with its own parameters, and it adds the CTC loss to the EEG objective. Without a control that uses a matched-capacity auxiliary branch with random phoneme targets (or some other non-phoneme task), you can't say the improvement comes from phoneme supervision rather than from regularization induced by any secondary gradient. The stress-test note has this right; this is load-bearing for the main claim, not a minor caveat.

Second, the abstract overstates. 'Outperforms previous methods in both modalities' is only true for the CB-2 config on the two clean test sets; CB-0 and CB-1 are actually worse than EEG2TEXT on unseen speech and unseen-both in Table III. And the 'unseen subject' set is contaminated because the target phoneme sequences appear in training, as the authors themselves note. That win shouldn't be headline material.

Third, the phoneme metric is teacher-forced top-k accuracy given previous phonemes. It's a defensible choice without an external LM, but it measures conditional decoding, not full-sequence decoding. The authors are transparent about this, so I'd call it minor.

Overall: the architecture is a reasonable contribution, the speech findings are plausible, and the paper deserves a serious referee. But the enhancement claim needs the random-target ablation, and the abstract needs to be reined in. I'd send it to review with that expectation.

Referee Report

2 major / 6 minor

Summary. The paper proposes a multi-task framework for listened-speech decoding from EEG, built on the authors' previous FESDE model. An auxiliary phoneme predictor (CB-0/1/2 variants) is attached to the EEG embeddings and trained with a CTC loss, while a gradient stop isolates the speech module. The paper reports that all three configurations outperform FESDE on speech reconstruction metrics (Table II) and that the best configuration outperforms an adapted EEG2TEXT baseline on phoneme-sequence top-k accuracy (Table III). It also presents a phoneme-group analysis of consonant and vowel categories. The central claim is that the auxiliary phoneme-prediction task enhances the primary speech-decoding task, enabling simultaneous speech and text output.

Significance. If the enhancement claim is established, the paper would show that a single EEG encoder can serve both speech waveform and phoneme decoding, turning a multi-modal output into a benefit rather than a trade-off. The strengths include the release of source code and speech samples, the use of a public dataset, consistent and large improvements in speech metrics, and a detailed phoneme-group analysis. However, the current experiments do not separate the effect of phoneme supervision from the added capacity and regularization of the auxiliary branch, and the phoneme evaluation is partly conditioned on true previous phonemes and includes a test condition with seen target sequences. These gaps prevent the significance of the reported gains from being fully supported. The phoneme-group analysis is a useful descriptive contribution that partially offsets these concerns.

major comments (2)

[Section III-B, Eq. (3), Table II] The attribution of the speech-decoding gains to phoneme supervision is confounded. OURS differs from FESDE by simultaneously (i) adding a parallel phoneme branch with extra parameters and (ii) adding alpha*L_CTC to the EEG-module objective in Eq. (3). Because a gradient stop prevents the speech module from influencing the EEG module, any speech gain must arise from changed EEG embeddings, but those embeddings are changed by the combined effect of the extra capacity and the CTC loss. No ablation trains FESDE with an equal-capacity auxiliary branch whose targets carry no phoneme information (e.g., randomized or permuted phoneme labels), nor a speech-only model with matched parameter count and training iterations. The CB-0 vs. CB-2 comparison shows that added capacity alone is not monotonically beneficial, but it does not rule out that any secondary gradient signal, rather than phoneme content specifically, regularizes the EEG encoder. This control is load-bearing for the paper's second contribution and should be added before the enhancement claim is accepted.
[Section IV-B, Table III, Fig. 2c] The reported phoneme metrics are conditional top-k accuracies computed given previous true phonemes (teacher forcing). This does not measure full phoneme-sequence decoding from EEG alone, because errors at earlier positions are not propagated; a model that exploits phonotactic regularities can score well even if it fails at the first phoneme. In addition, the 'unseen subject' test set shares target phoneme sequences with the training set, a fact the authors acknowledge when excluding it from Fig. 2c, yet the high numbers for that condition are still presented in Table III without a marker or caveat. The authors should report an unconditional or full-sequence metric (e.g., phoneme error rate with a language-model-free decoding protocol) and clearly separate the seen-phoneme condition from generalization claims.

minor comments (6)

[Abstract and Section IV] The statement that 'the proposed approach also outperforms previous methods in both modalities' is stronger than the data: in Table III, CB-1 is below EEG2TEXT-PT on the unseen-subject test set (51.07 vs. 53.38 top-1), and only CB-2 consistently outperforms the baseline. Please qualify the claim to the best configuration or specify the configuration being claimed.
[Section II-B and III-B] The notation for the CTC loss is inconsistent: Eq. (1) defines L_CTC, while Eq. (3) writes L_ctc. In addition, the variables z and \hat{z} in Eq. (1) are not explicitly defined in the text; please define the target and predicted phoneme sequences at that point.
[Section II-A1] There is a typo: 'ouputs' should be 'outputs'. Also, the sentence 'The EEG encoder takes preprocessed EEG signals as input and ouputs EEG embeddings' repeats the description of the EEG encoder; consider tightening the paragraph.
[Section III-A] The data-split description refers only to [16]; the number of sentences per held-out test set, the specific subject identifiers, and the degree of phoneme-sequence overlap between train and test are not specified. These details are needed to interpret Table III, especially for the unseen-subject condition.
[Section IV-B and Figure 2] The caption of Figure 2 contains typos ('T enseness', 'T op k Accuracy'), and the y-axis of panel (c) is labeled as a fraction (0.3-0.7) while the text calls it 'Top-3 accuracy (%)'. Please align the units and labels.
[Eq. (3) and Section III-B] The hyperparameter alpha = 0.3 is chosen heuristically and no sensitivity analysis is provided. Since the central claim concerns the effect of the auxiliary task, a small sweep around 0.3 (e.g., 0.1, 0.5) would increase confidence that the reported results are not tied to this specific value.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the claimed gains are empirical comparisons against separately trained baselines, not derivations from the model's own inputs.

full rationale

The paper's central claim is that adding a parallel phoneme predictor and the CTC objective improves both speech waveform decoding and phoneme sequence decoding. This is tested empirically in Table II and Table III against FESDE [16] and EEG2TEXT [8], where FESDE is retrained under the same setup rather than being used as a definitional component of the result. No evaluation metric (MCD, Mel-Corr, top-k accuracy) is fitted to the data, and no reported quantity is equal by construction to a training target or to a fitted parameter. The phoneme labels are obtained from the stimulus text via espeak G2P, which is standard supervised target construction rather than a circular prediction. The EEG and speech modules are adopted from the authors' own prior work [16], and the experimental setup and analysis pipeline are also taken from [16]; these are self-citations, but they are not load-bearing for the enhancement claim because the baseline is independently retrained and the comparison is an external empirical outcome. The manuscript itself flags a real limitation: in the unseen-subject phoneme test set, the target phoneme sequences were seen during training and the high accuracy 'may be due to overfitting.' A further experimental confound is that no control branch with random phoneme targets or matched extra capacity is provided, so the specific attribution of speech gains to phoneme content is not fully isolated. These are correctness and interpretation concerns, not circularity: they do not make any prediction equivalent to its input by definition. Therefore the appropriate circularity score is 0.

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

The paper introduces no new physical or mathematical entities. Its central claim depends on a heuristic loss weight, the assumption that the chosen dataset and preprocessing are valid, and the reliability of the G2P and forced-alignment tools for generating supervision and analysis targets.

free parameters (1)

alpha (CTC loss weight) = 0.3
Chosen heuristically in Eq. (3); controls the balance between EEG reconstruction and phoneme prediction, and the reported results depend on it. No sensitivity analysis is provided.

assumptions (4)

domain assumption The N400 dataset and the FESDE preprocessing pipeline are appropriate for listened speech decoding evaluation.
The entire experimental design is adopted from [16], including subjects, sentences, filtering, and test splits (Section III-A).
domain assumption espeak G2P produces phoneme targets that are correct and aligned with the perceived speech.
Phoneme labels are generated by espeak G2P (Section III-A) and used as ground truth for CTC training and evaluation; errors in G2P would directly affect reported accuracies.
standard math CTC loss is a valid objective for mapping EEG embeddings to phoneme sequences.
CTC is a standard sequence-to-sequence objective (Eq. 1) and is assumed to work here without adaptation.
domain assumption The Montreal Forced Aligner aligns phoneme intervals identically for ground-truth and reconstructed speech.
Section IV-D states 'the intervals of the ground-truth phonemes are assumed to be identical to those of the reconstructed ones', which is an assumption, not verified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Listened Speech Decoding from EEG via Parallel Phoneme Sequence Prediction." pith.science (2026). https://pith.science/paper/IONIODDO

@misc{pith2026250104844,
  author       = {Pith},
  title        = {Pith review of: Enhancing Listened Speech Decoding from EEG via Parallel Phoneme Sequence Prediction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IONIODDO}},
  note         = {Machine review of arXiv:2501.04844}
}

read the original abstract

Brain-computer interfaces (BCI) offer numerous human-centered application possibilities, particularly affecting people with neurological disorders. Text or speech decoding from brain activities is a relevant domain that could augment the quality of life for people with impaired speech perception. We propose a novel approach to enhance listened speech decoding from electroencephalography (EEG) signals by utilizing an auxiliary phoneme predictor that simultaneously decodes textual phoneme sequences. The proposed model architecture consists of three main parts: EEG module, speech module, and phoneme predictor. The EEG module learns to properly represent EEG signals into EEG embeddings. The speech module generates speech waveforms from the EEG embeddings. The phoneme predictor outputs the decoded phoneme sequences in text modality. Our proposed approach allows users to obtain decoded listened speech from EEG signals in both modalities (speech waveforms and textual phoneme sequences) simultaneously, eliminating the need for a concatenated sequential pipeline for each modality. The proposed approach also outperforms previous methods in both modalities. The source code and speech samples are publicly available.

Figures

Figures reproduced from arXiv: 2501.04844 by the authors.

**Figure 2.** (a) MCD (dB), (b) Mel-Corr (%), and (c) Top-3 accuracy (%) of each phoneme group with respect to the number of the conformer blocks. Better [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

28 extracted references · 18 canonical work pages

[1]

A high-performance speech neuroprosthesis,

F. R. Willett, E. M. Kunz, C. Fan, D. T. Avansino, G. H. Wilson, E. Y . Choi, F. Kamdar, M. F. Glasser, L. R. Hochberg, S. Druckmannet al., “A high-performance speech neuroprosthesis,” Nature, vol. 620, no. 7976, pp. 1031–1036, 2023

work page 2023
[2]

A high-performance neuroprosthesis for speech decoding and avatar control,

S. L. Metzger, K. T. Littlejohn, A. B. Silva, D. A. Moses, M. P. Seaton, R. Wang, M. E. Dougherty, J. R. Liu, P. Wu, M. A. Berger et al. , “A high-performance neuroprosthesis for speech decoding and avatar control,” Nature, vol. 620, no. 7976, pp. 1037–1046, 2023

work page 2023
[3]

Relating eeg to continuous speech using deep neural networks: a review

C. Puffay, B. Accou, L. Bollens, M. J. Monesi, J. Vanthornhout, T. Francart et al., “Relating eeg to continuous speech using deep neural networks: a review.” Journal of Neural Engineering , 2023

work page 2023
[4]

Knowledge- guided eeg representation learning,

A. Kommineni, K. Avramidis, R. Leahy, and S. Narayanan, “Knowledge- guided eeg representation learning,” in International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) , 2024

work page 2024
[5]

Bendr: using transform- ers and a contrastive self-supervised learning task to learn from massive amounts of eeg data,

D. Kostas, S. Aroca-Ouellette, and F. Rudzicz, “Bendr: using transform- ers and a contrastive self-supervised learning task to learn from massive amounts of eeg data,” Frontiers in Human Neuroscience , vol. 15, p. 653659, 2021

work page 2021
[6]

Maeeg: Masked auto-encoder for eeg representation learning,

H.-Y . S. Chien, H. Goh, C. M. Sandino, and J. Y . Cheng, “Maeeg: Masked auto-encoder for eeg representation learning,” in NeurIPS 2022 Workshop on Learning from Time Series for Health , 2022

work page 2022
[7]

Open vocabulary electroencephalography-to-text decoding and zero-shot sentiment classification,

Z. Wang and H. Ji, “Open vocabulary electroencephalography-to-text decoding and zero-shot sentiment classification,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 36, no. 5, 2022, pp. 5350–5358

work page 2022
[8]

Eeg2text: Open vocabulary eeg-to-text decoding with eeg pre-training and multi- view transformer,

H. Liu, D. Hajialigol, B. Antony, A. Han, and X. Wang, “Eeg2text: Open vocabulary eeg-to-text decoding with eeg pre-training and multi- view transformer,” arXiv preprint arXiv:2405.02165 , 2024

arXiv 2024

Show all 28 references

[9]

Dewave: Discrete encoding of eeg waves for eeg to text translation,

Y . Duan, C. Chau, Z. Wang, Y .-K. Wang, and C.-t. Lin, “Dewave: Discrete encoding of eeg waves for eeg to text translation,” Advances in Neural Information Processing Systems , vol. 36, 2024

2024
[10]

Decoding of the speech envelope from eeg using the vlaai deep neural network,

B. Accou, J. Vanthornhout, H. V . Hamme, and T. Francart, “Decoding of the speech envelope from eeg using the vlaai deep neural network,” Scientific Reports, vol. 13, no. 1, p. 812, 2023

2023
[11]

Convconcatnet: a deep convolutional neural network to reconstruct mel spectrogram from the eeg,

X. Xu, B. Wang, Y . Yan, H. Zhu, Z. Zhang, X. Wu, and J. Chen, “Convconcatnet: a deep convolutional neural network to reconstruct mel spectrogram from the eeg,” arXiv preprint arXiv:2401.04965 , 2024

2024 arXiv
[12]

Using articulated speech eeg signals for imagined speech decoding,

C. Bras, T. Patel, and O. Scharenborg, “Using articulated speech eeg signals for imagined speech decoding,” in Proc. INTERSPEECH 2024 – 20 th Annual Conference of the International Speech Communication Association, 2024

2024
[13]

De- coding speech perception from non-invasive brain recordings,

A. D ´efossez, C. Caucheteux, J. Rapin, O. Kabeli, and J.-R. King, “De- coding speech perception from non-invasive brain recordings,” Nature Machine Intelligence, vol. 5, no. 10, pp. 1097–1107, 2023

2023
[14]

WaveNet: A Generative Model for Raw Audio,

A. van den Oord, S. Dieleman, H. Zen, K. Simonyan, O. Vinyals, A. Graves, N. Kalchbrenner, A. Senior, and K. Kavukcuoglu, “WaveNet: A Generative Model for Raw Audio,” in Proc. 9th ISCA Workshop on Speech Synthesis Workshop (SSW 9) , 2016, p. 125

2016
[15]

Cross-attention-guided wavenet for mel spectrogram reconstruction in the icassp 2024 auditory eeg challenge,

Y . Fang, H. Li, X. Zhang, F. Chen, and G. Gao, “Cross-attention-guided wavenet for mel spectrogram reconstruction in the icassp 2024 auditory eeg challenge,” in 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing Workshops (ICASSPW) , 2024, pp. 7–8

2024
[16]

Toward fully-end-to-end listened speech decoding from eeg signals,

J. Lee, A. Kommineni, T. Feng, K. Avramidis, X. Shi, S. Kadiri, and S. Narayanan, “Toward fully-end-to-end listened speech decoding from eeg signals,” in Proc. Interspeech 2024 , 2024

2024
[17]

Efficiently modeling long sequences with structured state spaces,

A. Gu, K. Goel, and C. Re, “Efficiently modeling long sequences with structured state spaces,” in Proc. ICLR, 2021

2021
[18]

Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech,

J. Kim, J. Kong, and J. Son, “Conditional variational autoencoder with adversarial learning for end-to-end text-to-speech,” in International Conference on Machine Learning . PMLR, 2021, pp. 5530–5540

2021
[19]

Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis,

J. Kong, J. Kim, and J. Bae, “Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis,” Advances in Neural Information Processing Systems , vol. 33, pp. 17 022–17 033, 2020

2020
[20]

Attention is all you need,

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. u. Kaiser, and I. Polosukhin, “Attention is all you need,” in Advances in Neural Information Processing Systems , I. Guyon, U. V . Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garn...

2017
[21]

Density estimation using real NVP,

L. Dinh, J. Sohl-Dickstein, and S. Bengio, “Density estimation using real NVP,” in International Conference on Learning Representations ,
[22]

Conformer: Convolution- augmented Transformer for Speech Recognition,

A. Gulati, J. Qin, C.-C. Chiu, N. Parmar, Y . Zhang, J. Yu, W. Han, S. Wang, Z. Zhang, Y . Wu, and R. Pang, “Conformer: Convolution- augmented Transformer for Speech Recognition,” in Proc. Interspeech 2020, 2020, pp. 5036–5040

2020
[23]

Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks,

A. Graves, S. Fern ´andez, F. Gomez, and J. Schmidhuber, “Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks,” in Proceedings of the 23rd International Conference on Machine Learning , ser. ICML ’06. New York, NY , USA: As...

2006
[24]

Evoking the n400 event-related potential (erp) component using a publicly available novel set of sentences with semantically incongruent or congruent eggplants (endings),

K. K. Toffolo, E. G. Freedman, and J. J. Foxe, “Evoking the n400 event-related potential (erp) component using a publicly available novel set of sentences with semantically incongruent or congruent eggplants (endings),” Neuroscience, vol. 501, pp. 143–158, 2022

2022
[25]

Decoupled weight decay regularization,

I. Loshchilov and F. Hutter, “Decoupled weight decay regularization,” in International Conference on Learning Representations , 2019

2019
[26]

Mel-cepstral distance measure for objective speech qual- ity assessment,

R. Kubichek, “Mel-cepstral distance measure for objective speech qual- ity assessment,” in Proceedings of IEEE Pacific Rim Conference on Communications Computers and Signal Processing , vol. 1, 1993, pp. 125–128 vol.1

1993
[27]

Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi,

M. McAuliffe, M. Socolof, S. Mihuc, M. Wagner, and M. Sonderegger, “Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi,” in Proc. Interspeech 2017 , 2017, pp. 498–502

2017
[2017]

Available: https://openreview.net/forum?id=HkpbnH9lx

[Online]. Available: https://openreview.net/forum?id=HkpbnH9lx

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

A high-performance speech neuroprosthesis,

F. R. Willett, E. M. Kunz, C. Fan, D. T. Avansino, G. H. Wilson, E. Y . Choi, F. Kamdar, M. F. Glasser, L. R. Hochberg, S. Druckmannet al., “A high-performance speech neuroprosthesis,” Nature, vol. 620, no. 7976, pp. 1031–1036, 2023

work page 2023

[2] [2]

A high-performance neuroprosthesis for speech decoding and avatar control,

S. L. Metzger, K. T. Littlejohn, A. B. Silva, D. A. Moses, M. P. Seaton, R. Wang, M. E. Dougherty, J. R. Liu, P. Wu, M. A. Berger et al. , “A high-performance neuroprosthesis for speech decoding and avatar control,” Nature, vol. 620, no. 7976, pp. 1037–1046, 2023

work page 2023

[3] [3]

Relating eeg to continuous speech using deep neural networks: a review

C. Puffay, B. Accou, L. Bollens, M. J. Monesi, J. Vanthornhout, T. Francart et al., “Relating eeg to continuous speech using deep neural networks: a review.” Journal of Neural Engineering , 2023

work page 2023

[4] [4]

Knowledge- guided eeg representation learning,

A. Kommineni, K. Avramidis, R. Leahy, and S. Narayanan, “Knowledge- guided eeg representation learning,” in International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC) , 2024

work page 2024

[5] [5]

Bendr: using transform- ers and a contrastive self-supervised learning task to learn from massive amounts of eeg data,

D. Kostas, S. Aroca-Ouellette, and F. Rudzicz, “Bendr: using transform- ers and a contrastive self-supervised learning task to learn from massive amounts of eeg data,” Frontiers in Human Neuroscience , vol. 15, p. 653659, 2021

work page 2021

[6] [6]

Maeeg: Masked auto-encoder for eeg representation learning,

H.-Y . S. Chien, H. Goh, C. M. Sandino, and J. Y . Cheng, “Maeeg: Masked auto-encoder for eeg representation learning,” in NeurIPS 2022 Workshop on Learning from Time Series for Health , 2022

work page 2022

[7] [7]

Open vocabulary electroencephalography-to-text decoding and zero-shot sentiment classification,

Z. Wang and H. Ji, “Open vocabulary electroencephalography-to-text decoding and zero-shot sentiment classification,” in Proceedings of the AAAI Conference on Artificial Intelligence , vol. 36, no. 5, 2022, pp. 5350–5358

work page 2022

[8] [8]

Eeg2text: Open vocabulary eeg-to-text decoding with eeg pre-training and multi- view transformer,

H. Liu, D. Hajialigol, B. Antony, A. Han, and X. Wang, “Eeg2text: Open vocabulary eeg-to-text decoding with eeg pre-training and multi- view transformer,” arXiv preprint arXiv:2405.02165 , 2024

arXiv 2024