The ASME-speller: 30-class auditory brain-computer interface speller using stream segregation and the QWERTY layout.

Overview

1. Graduate School of Engineering and Science, Shibaura Institute of Technology, Tokyo, Japan
2. College of Engineering, Shibaura Institute of Technology, Tokyo, Japan

Abstract

Introduction: This study presents the ASME-speller, a novel 30-class auditory brain-computer interface (BCI) speller system that combines auditory stream segregation with the familiar QWERTY keyboard layout to facilitate intuitive and visionfree communication.

Methods: In the ASME-speller, three distinct auditory streams are presented simultaneously, each corresponding to a row on the QWERTY keyboard. The low-, middle-, and high-frequency streams represent the bottom, middle, and top rows, respectively. Within each stream, alphabet letters and selected symbols are repeatedly presented as spoken voice stimuli. Users are instructed to focus exclusively on the stream corresponding to the row containing the target letter and to selectively attend to that letter within the stream. By leveraging the QWERTY layout and auditory stream segregation, the proposed approach enables users to restrict their attentional focus to a subset of letters by directing selective attention to auditory streams, while the mapping between QWERTY rows and stream pitch facilitates intuitive letter selection. We conducted online experiments with ten healthy participants to evaluate system performance.

Results: The ASME-speller achieved an average classification accuracy of 0.76 and an average information transfer rate (ITR) of 2.16 bits/min. Excluding one participant whose EEG data contained excessive artifacts, these values improved to 0.84 and 2.40 bits/min, respectively. Post-hoc analyses further examined the effects of preprocessing parameters, classification pipelines, and early stopping strategies. Among four pipelines tested, a linear discriminant analysis (LDA) combined with dynamic stopping demonstrated the most robust performance across participants (accuracy of 0.80 and ITR of 4.76 bits/min). For the best participant, a deep learning model (EEGNet4,2) with dynamic stopping achieved accuracy of 1.0 with ITR of 14.44 bits/min.

Discussion: Compared to previous auditory BCI spellers, the ASME-speller demonstrates performance comparable to existing systems, while offering advantages in terms of simplicity, requiring only standard headphones and no visual support. These findings demonstrate the feasibility of the ASME-speller and pave the way toward practical auditory BCI applications for communication.

Reproduced under the paper's license (CC BY), from the paper cited above.

Code

The paper links to its data, not to its authors' code: see the Data section.

Tracing map

A tracing map links a paper to the code its authors published: this paper has none, so it has no map.

Data

Datasets cited

• doi:10.7910/dvn/tyrcwl, at the source; found in “Data availability statement”

Data availability statement

All relevant data are publicly available on the Harvard Dataverse repository (https://doi.org/10.7910/DVN/TYRCWL).

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

• Funding: added Japan Society for the Promotion of Science: 23K11811

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 2 authors, 8 keywords, 65 references.

Cite

This paper

Kojima, S., & Kanoh, S. (2026). The ASME-speller: 30-class auditory brain-computer interface speller using stream segregation and the QWERTY layout. Frontiers in human neuroscience, 20, 1807535. https://doi.org/10.3389/fnhum.2026.1807535

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.3389/fnhum.2026.1869918 [code]

Single-subject auditory ERP-BCI performance enhancement in ALS via an AI coding assistant prompt.

Journal: Frontiers in human neuroscience

In common: EEG, 9 references

[2] doi:10.3389/fnhum.2026.1763477

Toward precision EEG: assessing the reliability of individual-level ERPs across EEG systems.

Journal: Frontiers in human neuroscience

In common: EEG, 4 references

[3] doi:10.3390/s26113310

A Robust Multi-Branch CNN-LSTM Architecture for Cross-Subject Motor Imagery Classification.

Journal: Sensors (Basel, Switzerland)

In common: EEG, 5 references

[4] doi:10.1186/s12984-026-02041-3 [code]

Mental tasks induce common modulations of oscillations in cortex and spinal cord.

Journal: Journal of neuroengineering and rehabilitation

In common: EEG, 4 references

[5] doi:10.3389/fnhum.2026.1832475

Optimization of stimulus color for peripheral SSVEP-based brain-computer interfaces.

Journal: Frontiers in human neuroscience

In common: EEG, 3 references

[6] doi:10.3389/fnhum.2026.1811759

MCFANet: a multi-class fusion attention network for motor imagery EEG classification.

Journal: Frontiers in human neuroscience

In common: EEG, 4 references

[7] doi:10.3390/s26134045

Brain Signal for Secure EEG Biometric Authentication: A Comprehensive Survey.

Journal: Sensors (Basel, Switzerland)

In common: methods / tools, EEG, 3 references

[8] doi:10.1016/j.dib.2026.113148

An ERP dataset for multi-information identity authentication.

Journal: Data in brief

In common: methods / tools, EEG, 3 references

[9] doi:10.3390/brainsci16080873

Evaluating Validation Strategies in Motor Imagery EEG: A Full-Cohort GAF-PLV Analysis and Matched Sensitivity Study.

Journal: Brain sciences

In common: methods / tools, EEG, 3 references

[10] doi:10.3390/bios16070394

Hybrid Edge-Cloud Asymmetric Analytics for Portable Multimodal BCI Biosensors.

Journal: Biosensors

In common: methods / tools, EEG, 3 references

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.