Contextual Awareness Resolves Noisy Non-Invasive Neural Speech Signals
Aiden Smith · We Interviewed the New Thiel Fellow Dropouts · May 28, 2025 · at 1:50:30
Aiden Smith, founder of Interphases and former Neuralink engineer, explains how language models and situational context make noisy neural speech signals usable.
0:00 / 0:22exact quote · 22.1s
▶ Watch the full episode on YouTube → 720p mp4 · rendered on demand · StarZero watermark
“There's a really good story to be told about kind of doing this for speech and getting really terrible speech. I have no illusions as a neurotechnologist that you will get really terrible speech out of this. But the key thing is that this does not matter because the context is what matters. And by informing this really awful speech on the context of what's around you and like your intention, the situation, you may be able to actually get the underlying part of it that matters.”
quote is from the automated transcript, cleaned for reading: filler sounds and stutters are removed, nothing is rephrased. names can be misheard (the analysis reads context, assessments check outside sources). how →
More from Aiden Smith
Voice Interfaces Fail for Ambient AI Due to Social Awkwardness
“I actually don't like voice as a medium that much for most things. If it actually wants to integrate into your life effortlessly, I don't want to be in the middle of conversation with you and be like, what's the guy on the left's name. It's super awkward. It's…”