We are conducting audio chat testing by streaming using Wowza. At this time, the speaker output is input to the microphone and the sound is generated. Is there a way to handle it?
An audio feedback loop is a common problem when dealing with video-calling or -conferencing. This isn’t specific to Wowza, it can happen with any software. The best solution is to use a headset or otherwise avoid that the sound from the speakers can be picked up by the microphone.
If you’re using WebRTC there are some tricks that can reduce it, like enabling echo cancelation. Also, make sure that all participants in the call are using the same audio sample rate (not sure if it’s even possible to use mixed sample rates in a single call, but worth checking), because a mix of 44.1KHz and 48KHz will cause an echo.
Generally, googling for “webrtc feedback loop” will give you lots of search results with likely some good tips.