Related Experiment Video
Updated: May 10, 2026
05:38
Interaction between Phonological and Semantic Processes in Visual Word Recognition using Electrophysiology
Published on: June 29, 2021
ASEAF: attention-SincNet driven EEG-audio fused target speaker extraction network
Yuhang Yang1, Yuan Liao2, Qiushi Han1
1College of electronic and optical engineering & college of flexible electronics (future technology), Nanjing University of Posts and Telecommunications, Jiangsu 210023, People's Republic of China.
Biomedical Physics & Engineering Express
|May 8, 2026
Summary
This study introduces ASEAF, an EEG-based model for extracting target speech in noise by fusing brain and audio signals. It significantly improves speech reconstruction for hearing aid solutions and brain-computer interfaces.
Area of Science:
- Neuroscience
- Signal Processing
- Machine Learning
Background:
- Selective auditory attention is challenging in noisy environments.
- Existing methods struggle with precise target speaker extraction.
- Neural decoding offers potential for advanced hearing assistance.
Purpose of the Study:
- To develop an EEG-based model (ASEAF) for target speaker extraction in noisy conditions.
- To improve speech reconstruction using simultaneous EEG and audio signal processing.
- To enhance hearing aid solutions and brain-computer interfaces (BCIs).
Main Methods:
- Utilized a Convolutional Neural Network (CNN) and self-attention for EEG spatio-temporal features.
- Employed SincNet for frequency-aware audio feature extraction.
- Integrated a dual-path LSTM and CNN decoder for cross-modal fusion and waveform reconstruction.
Main Results:
- ASEAF demonstrated superior performance over state-of-the-art models on multiple datasets.
- Achieved an average improvement of 11.5% in scale-invariant signal-to-distortion ratio (SI-SDRi).
- Showcased effective cross-modal interaction insights between EEG and audio signals.
Conclusions:
- ASEAF provides a robust solution for selective auditory attention in complex acoustic environments.
- The model offers significant advancements for hearing aid technology.
- This research contributes to the progress of brain-computer interfaces through neural signal processing.
