End-to-End Intracortical Speech Decoding from Neural Activity
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913157934481408 |
|---|---|
| author | Khanday, Owais Mujtaba Gonzalez-Lopez, Jose A. Ouellet, Marc Galdon, Alberto Granados, Gonzalo Olivares |
| author_facet | Khanday, Owais Mujtaba Gonzalez-Lopez, Jose A. Ouellet, Marc Galdon, Alberto Granados, Gonzalo Olivares |
| contents | Current high-performing intracortical speech neuroprostheses achieve low word error rates but typically rely on external language models during inference, increasing memory, computation, and latency. In this work, we investigate whether meaningful character-level decoding is achievable without such models. We propose an end-to-end Conformer-based neural decoder trained directly on intracortical recordings from a participant with amyotrophic lateral sclerosis (ALS). Without any external language model, the system achieves a character error rate (CER) of 23.80\% on held-out validation data. Analysis shows that performance variability is driven by inter-session signal degradation, while dominant errors arise from incorrect word boundary segmentation. These results demonstrate that effective character-level decoding is possible in a fully end-to-end framework, providing a strong neural signal for downstream linguistic processing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_24313 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | End-to-End Intracortical Speech Decoding from Neural Activity Khanday, Owais Mujtaba Gonzalez-Lopez, Jose A. Ouellet, Marc Galdon, Alberto Granados, Gonzalo Olivares Computation and Language Human-Computer Interaction Current high-performing intracortical speech neuroprostheses achieve low word error rates but typically rely on external language models during inference, increasing memory, computation, and latency. In this work, we investigate whether meaningful character-level decoding is achievable without such models. We propose an end-to-end Conformer-based neural decoder trained directly on intracortical recordings from a participant with amyotrophic lateral sclerosis (ALS). Without any external language model, the system achieves a character error rate (CER) of 23.80\% on held-out validation data. Analysis shows that performance variability is driven by inter-session signal degradation, while dominant errors arise from incorrect word boundary segmentation. These results demonstrate that effective character-level decoding is possible in a fully end-to-end framework, providing a strong neural signal for downstream linguistic processing. |
| title | End-to-End Intracortical Speech Decoding from Neural Activity |
| topic | Computation and Language Human-Computer Interaction |
| url | https://arxiv.org/abs/2605.24313 |