1 post • Page 1 of 1
I have a project I'm trying to implement where I need a Raspberry Pi to listen to a live audio stream (provided via a Behringer USB Audio interface) and close a relay when it detects a change in speaker in a multi-speaker stream. I can find a ton of information on Automated Speech Recognition and Speech to Text using a Raspberry Pi (which I've already addressed), but I'm not finding much regarding live speaker identification. Has anyone seen anything that is doing anything similar? The average Speech to Text tends to output just a wall of text and I'd like to use the relay closure to insert a symbol in that text to indicate the change in speaker to the end user and hopefully make the text easier to comprehend. Thanks.