I’m testing voice conversations by accessing the Live API with a WebSocket in a c++ based environment (unreal).
It receives a piece of generated voice data in real time and is playing the sound in a streaming manner.
However, the playback speed is faster than the data reception speed, resulting in a voice lag (stuttering) in the middle.
The first few words play well, but if the words get longer, they always get into trouble.
I’m starting playback a one second after first receiving the voice, but it doesn’t solve it.
Has anyone had the same problem as me, solved it?