dimanshu
dimanshu
@rookie-may did u solve it?
@xjj210130 did u solve it ?
@MatthewScholefield is it necessary to put wake word at the end ? what are the changes I have to do so that can put wake word anywhere in the audio....
@MatthewScholefield thanks last week i have completed my model for emotion detection for calls, now Im working for keyword detection in calls . i have keyword data( generated from tts...
when i check 3 second file with my model it is giving me this result . 0.1399751031411294 0.1399751031411294 0.1399751031411294 0.09472132607955966 0.15695710929573797 0.15695710929573797 0.1257667514553736 0.0684353295182932 0.0684353295182932 0.011264940791642497 0.0024931440806811476 0.0024931440806811476 0.0005406289277940763 0.0005406289277940763...
yes im testing manually if it works then I will make api
@MatthewScholefield can i train multiple keyword at the same time ?and what are the changes i need to do for that ?
now i'm using mfcc decreasing . no error in mfcc
no bro still struggling