I try to use EnergyDetector as voice activity detector. I have tried a lot of different settings. For now as far as I understand EnergyDetector uses only energy information as input. So when I have some file without speech it tries to get maximum energy segment. After EnergyDetector makes .lbl file with marked regions. And also how many percent of the file is selected with the speech. But actually where is now speech in this file. Am I right?
I try to use EnergyDetector as voice activity detector. I have tried a lot of different settings. For now as far as I understand EnergyDetector uses only energy information as input. So when I have some file without speech it tries to get maximum energy segment. After EnergyDetector makes .lbl file with marked regions. And also how many percent of the file is selected with the speech. But actually where is now speech in this file. Am I right?