Having proper sub-vocalization interfaces would be so cool! A 19-channel EEG does seem a bit invasive though (from a social aspect, I understand that it's used here in a medical context).

Subvocalization can be decoded from facial muscles.

MIT’s AlterEgo project does that. It’s been spun off into a private company (with no shipping product yet.) Discussed here, with 130 comments, 11 months ago: https://news.ycombinator.com/item?id=45174125