Here’s the Merc story today about how a UC Berkeley computer science student figured out how to make audio recordings of keyboard strokes that deciphered which words were being typed.
She used a cheap microphone plugged into a laptop running generic speech recognition software. “It’s possible that intelligence agencies are using this method,” said her professor, “and it’s possible that some bad guys already have this.”
The audio problem: Learn how new cloud-based API solutions are solving imperfect, frustrating audio in video conferences. Access here