Prolific science vlogger and author Steve Mould explained how acoustic cameras see sounds with dynamic range, similar to how a traditional camera sees light. Mould visited an anechoic chamber to learn more about how the acoustic camera would respond to him speaking while playing the ukulele.
What we have here is an issue of dynamic range. If you have your camera set to automatic exposure which your smartphone is by default, and you shoot a video of a dim light, then turn on a much brighter light, the camera will switch between showing you lots of detail of the dim light to showing you lots of detail of the bright light. And the same thing is happening here with the ukulele.
As it turned out, the camera alternated its response to whichever sound was the loudest at the time.
The ukulele is louder than my voice. So when the ukulele is being played, the acoustic camera dials down its sensitivity and you can no longer see my voice. But when I stop strumming the ukulele, the camera dials up.