analiza aiqbench
This research proposes "Brain2Semantics2Text," a novel framework that bypasses the low signal-to-noise ratio of non-invasive brain recordings by targeting high-level semantic representations instead of acoustic features. By mapping MEG responses into an intermediate semantic manifold before reconstructing text, the method achieves improved sentence-level decoding without requiring word-level alignment. This approach represents a significant methodological advance in machine learning for neural interface applications, offering a more robust pathway for practical non-invasive speech restoration.
The Semantic Bottleneck: Leveraging Semantic Representations for Non-Invasive Speech Decoding
arXiv:2609.10296v1 Announce Type: cross Abstract: Non-invasive speech decoding remains constrained by the low signal-to-noise ratio of neural recordings, which makes fine-grained reconstruction of phonemes or individual words difficult. Motivated by neuroscientific evidence that high-level semantic representations are distributed across cortical regions and evolve over slower temporal scales, we hypothesize that semantic content may provide a more suitable target for non-invasive decoding than low-level acoustic or lexical features. We introduce Brain2Semantics2Text, a method that reconstructs text through an intermediate semantic embedding space. Our model maps sentence-level MEG responses into a semantic manifold and then inverts the predicted embeddings into natural language. This semantic bottleneck enables recovery of high-level meaning without word-level alignment. We describe the core principles of the approach, its implementation, and the strategies used to mitigate the challenges of learning a reliable neural-to-semantic mapping. Finally, we compare against prior non-invasive Brain2Text methods and show improved sentence-level results.
Abstract:Non-invasive speech decoding remains constrained by the low signal-to-noise ratio of neural recordings, which makes fine-grained reconstruction of phonemes or individual words difficult. Motivated by neuroscientific evidence that high-level semantic representations are distributed across cortical regions and evolve over slower temporal scales, we hypothesize that semantic content may provide a more suitable target for non-invasive decoding than low-level acoustic or lexical features. We introduce Brain2Semantics2Text, a method that reconstructs text through an intermediate semantic embedding space. Our model maps sentence-level MEG responses into a semantic manifold and then inverts the predicted embeddings into natural language. This semantic bottleneck enables recovery of high-level meaning without word-level alignment. We describe the core principles of the approach, its implementation, and the strategies used to mitigate the challenges of learning a reliable neural-to-semantic mapping. Finally, we compare against prior non-invasive Brain2Text methods and show improved sentence-level results.
Computation and Language (cs.CL); Machine Learning (cs.LG)
arXiv:2609.10296 [cs.CL]
(or arXiv:2609.10296v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.10296 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 241 cuvinte
Subiecte conexe
From Tokens to Faces: Investigating Discrete Speech Representations for 3D Facial Animation
arXiv · 12 iun. 2026
Brain-to-Language Decoding: Tasks, Signals, Methods, Evaluation, Practical Use and Beyond
arXiv · 24 sept. 2026
Cross-Subject Semantic Decoding with Shared-Space Alignment for Generalized Neural Representation Learning
arXiv · 23 iul. 2026
Is Text All You Need? Text as a Universal Information Bottleneck for Speech LLMs
arXiv · 16 oct. 2026