LightBeam: An Accurate and Memory-Efficient CTC Decoder for Speech Neuroprostheses
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912966221234176 |
|---|---|
| author | Feghhi, Ebrahim Hu, Junlin Hadidi, Nima Kao, Jonathan C. |
| author_facet | Feghhi, Ebrahim Hu, Junlin Hadidi, Nima Kao, Jonathan C. |
| contents | A promising pathway for restoring communication in patients with dysarthria and anarthria is speech neuroprostheses, which directly decode speech from cortical neural activity. Two benchmarks, Brain-to-Text '24 and '25, released intracranial recordings from patients with dysarthria along with a baseline algorithm trained with Connectionist Temporal Classification (CTC). Despite significant innovation on these benchmarks, all leading published prior work relies on a WFST-based CTC decoder that requires ${\sim}$320 GB of RAM. These memory requirements limit accessibility for both patients and researchers. Here, we propose LightBeam, a non-WFST based CTC decoder that requires only ${\sim}$10 GB of RAM and achieves state-of-the-art performance on both benchmarks. LightBeam achieves this by integrating an LLM into the beam-search process via delayed fusion, obviating the prior need for using a large N-gram LM. LightBeam is implemented in Python and is open-source. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_14002 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LightBeam: An Accurate and Memory-Efficient CTC Decoder for Speech Neuroprostheses Feghhi, Ebrahim Hu, Junlin Hadidi, Nima Kao, Jonathan C. Human-Computer Interaction Sound A promising pathway for restoring communication in patients with dysarthria and anarthria is speech neuroprostheses, which directly decode speech from cortical neural activity. Two benchmarks, Brain-to-Text '24 and '25, released intracranial recordings from patients with dysarthria along with a baseline algorithm trained with Connectionist Temporal Classification (CTC). Despite significant innovation on these benchmarks, all leading published prior work relies on a WFST-based CTC decoder that requires ${\sim}$320 GB of RAM. These memory requirements limit accessibility for both patients and researchers. Here, we propose LightBeam, a non-WFST based CTC decoder that requires only ${\sim}$10 GB of RAM and achieves state-of-the-art performance on both benchmarks. LightBeam achieves this by integrating an LLM into the beam-search process via delayed fusion, obviating the prior need for using a large N-gram LM. LightBeam is implemented in Python and is open-source. |
| title | LightBeam: An Accurate and Memory-Efficient CTC Decoder for Speech Neuroprostheses |
| topic | Human-Computer Interaction Sound |
| url | https://arxiv.org/abs/2603.14002 |