4 ms·

Would it be difficult to make a program that sits in the audio stack (probably on Windows) and tries to run voice recognition on everything coming over the line

by foota 9y ago

Would it be difficult to make a program that sits in the audio stack (probably on Windows) and tries to run voice recognition on everything coming over the line and then renders anything that sounds like speech at the bottom of the screen?

Garfgon 9y ago
Yes. Youtube tries (tried?) to autocaption videos, and the results were infamously bad.
rhizome 9y ago
Yes. Voice recognition for arbitrary voices, accents, and lexicons is not as reliable as you imagine, much less in real-time.