Summary
People who cannot speak might regain communication through devices that read brain activity related to speech and convert it into text or synthesized voice. The authors show that such brain-computer interfaces (BCIs) are complex systems combining brain signals, recording devices, machine learning models, and user interaction rather than simple translators. They emphasize that future BCIs must be designed for reliability, ease of use, and user control, balancing technical trade-offs and ethical concerns. The paper reviews the current state of speech BCIs and outlines the multidisciplinary work needed to turn early successes into practical communication tools.
What this means in practice
- •For neural engineering teams: Integrate recording hardware and adaptive decoders to build more robust speech BCIs that can be calibrated and updated alongside user learning.
- •For rehabilitation device developers: Develop communication aids for people with speech impairment using insights about neural speech signals and decoding challenges across sessions.$Commercial implications: Enables creation of communication neuroprostheses for patients by informing better design of devices translating brain activity to speech output.
A survey. It maps existing work.
Authors
Moein Khajehnejad, Forough Habibollahi, Tommaso Boccato, Margarida Sousa, Michal Olak, Francesco Jamal Sheiban, Matteo Ferrante
Abstract
Speech brain-computer interfaces (BCIs) aim to restore communication by transforming neural activity related to speech, language, or communicative intent into external outputs such as text, synthesized voice, or avatar control. Recent advances in intracortical and electrocorticographic recording, deep sequence models, and language-model-assisted decoding have enabled rapid progress, including high-performance attempted-speech decoding and increasingly naturalistic speech synthesis. Yet these achievements also reveal that speech BCIs are not simply neural-to-text decoders. They are adaptive clinical systems in which neural representations, recording hardware, decoding architectures, language priors, feedback, and user learning interact over time. Here, we synthesize speech BCI research from a system-level perspective. We first examine the neural substrates of speech and language, emphasizing their hierarchical, distributed, temporally structured, and non-stationary organization. We then examine recording and decoding choices, closed-loop adaptation, evaluation, clinical translation, and ethics. Across these domains, we highlight recurring trade-offs between signal resolution and invasiveness, low-level motor and high-level semantic targets, decoder accuracy and user agency, and language-model fluency and faithful neural evidence. We argue the next generation of speech BCIs should be evaluated not only by offline accuracy, but also by robustness across sessions, calibration burden, latency, uncertainty, usability, and safeguards against unintended decoding. By reframing speech BCIs as adaptive, user-centred systems, we outline the interdisciplinary priorities spanning speech neuroscience, neural engineering, machine learning, clinical practice, and neuroethics needed to move from proof-of-concept decoding toward reliable, expressive, and controllable communication neuroprostheses.