All Topics
All Topics
Technology
Technology
Design
Design
Programming
Programming
Science
Science
News
News
Gaming
Gaming
Entertainment
Entertainment
Business
Business
Finance
Finance
Sports
Sports
Health
Health
Food
Food
Travel
Travel
Art
Art
Music
Music
Books
Books
Education
Education
Politics
Politics
Personal
Personal
No algorithm. No AI slop. No ads. Just RSS. Pro-human. Indie writers. Real journalism. Open web. Chronological. Hand toasted.

Research Team Collects 10,000 Hours of Neuro-Language Data for Thought-to-Text Models

By

nee1r

5mo ago· 39 min readenInsight

Summary

A research team has collected approximately 10,000 hours of neuro-language data from thousands of individuals over six months, claiming it to be the world's largest dataset of its kind. The data collection supports their work on training thought-to-text models that decode semantic content from noninvasive neural data, with the goal of enabling brain-computer interfaces for communication. The article discusses their methodology, challenges with existing small datasets, and presents zero-shot examples of their model's capabilities.

Key quotes

· 4 pulled
Over the last 6 months, we collected ~10k hours of data across thousands of unique individuals.
As far as we know, this is the largest neuro-language dataset in the world.
We train thought-to-text models. That is, we train models to decode semantic content from noninvasive neural data.
Here are some entirely zero-shot examples:
Snippet from the RSS feed
Over the last 6 months, we collected ~10k hours of data across thousands of unique individuals. As far as we know, this is the largest neuro-language dataset in the world.[1]See here, here, here, here, and here (discussion only, no data available) for som

You might also wanna read