←
Erin Kunz31 posts
Erin Kunz avatar

Erin Kunz

@pearlsandpython

Postdoc @ Stanford Neural Prosthetics Translational Lab | NeuroEngineer working on Speech BCI 🧠 💻

◷ Joined October 2017▧ 9 Photos and videos
Neuralink avatar
⟳ Erin Kunz retweeted
Neuralink@neuralink·Sep 16

A beautiful surprise

0:23
Benyamin Abramovich Krasa avatar
⟳ Erin Kunz retweeted
Benyamin Abramovich Krasa@BenyaminAK_BCI·Jun 18

Cool to see the Brain to text '24 and '25 contest results to the test for online decoding!

🧠🗣️ In speech BCIs decoding text from brain activity, accuracy shapes how reliably people with paralysis can communicate. New preprint on cloud-based ensembles as a strategy for improving speech BCI accuracy (link below) Watch the BCI think twice👇1/10

0:45
Erin Kunz avatar
Erin Kunz@pearlsandpython·May 11

Excited to see this out! This project includes some of my earliest PhD experiments and findings with participant T12. Huge congrats to @BenyaminAK_BCI and the rest of the team for building on it and getting it across the finish line!

🧠🗣️ Ever stumbled over your words, realizing only after speaking them outloud? Your brain might not have followed the plan. New preprint on the neural ensemble organization of speech motor plans and what it means for speech BCIs doi.org/10.64898/2026.04.27.… 1/10

0:33
Alisa Levin avatar
⟳ Erin Kunz retweeted
Alisa Levin@alisadlevin·Jan 15

Excited to share our new work on cross-brain transfer of speech and handwriting BCIs! Some studies have required 10+ days of training data for peak BCI performance. Can pooling neural data from other users speed things up? 1/6 biorxiv.org/content/10.64898…

Post attachment
The Naked Scientists avatar
⟳ Erin Kunz retweeted
The Naked Scientists@NakedScientists·15 Aug 2025

News podcast: @pearlsandpython @Stanford on inner voice; @SamuelDickenUK @ucl on ultra-processed foods; Solomon Hirsch @imperialcollege on Mars; and Katie Shanks @Renewables_UoE and Sophie Gledhill @FraunhoferISE on butterfly-inspired solar energy. thenakedscientists.com/podca…

Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

Our findings show that inner speech could be a powerful new modality for BCIs—but also highlight the importance of privacy & agency. By decoding imagined sentences and creating safeguards, we aim to move toward communication neuroprosthetics that are empowering in all aspects.

Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

To address this, we propose two simple yet effective strategies—a new way to train an attempted speech BCI to ignore inner speech, and a password-protection system that prevents inner speech from being decoded unless the user first imagines the password

Post attachment
Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

But there’s a challenge: if inner speech lives in motor cortex, could a BCI accidentally decode something a user only intended to think, not say? In controlled cognitive tasks, like silently counting shapes in a grid, we found BCIs could indeed pick up number words.

Post attachment
Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

Inner speech was highly correlated with attempted speech. But we discovered a unique “motor-intent” dimension that separates them—with ALS who was completely anarthric—suggesting attempted speech remains a viable BCI driver even in near locked-in states.

Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

For people who’ve lost the ability to speak, a BCI decoding inner speech might be more comfortable by bypassing physical effort while still enabling fluent communication. Here T16 uses an attempted speech BCI :speaking:

0:08
Erin Kunz avatar
Erin Kunz@pearlsandpython·15 Aug 2025

We investigated inner speech in the motor cortex—finding a robust representation that enabled real-time decoding of imagined sentences with error rates as low as 14% for a 50-word and 26% for a 125,000-word vocabulary

0:52
Sergey Stavisky avatar
⟳ Erin Kunz retweeted
Sergey Stavisky@SergeyStavisky·6 Oct 2024

Interested in speech neuroprostheses? Come to the #SfN24 minisymposium organized by @HerffC and I. We have a fantastic lineup chosen to survey the many directions this field can go in: different recording modalities, languages, and behaviors (e.g. attempted vs. imagined speech).

Post attachment