What the Max chunk size to be sent on WSS Audio to Video API

Hello, we are using the WSS Audio to Video API (https://docs.heygen.com/reference/audio-to-video-api), and we would like to know what is the max amount of audio bytes can be sent by chunk.
We are currently sending around 120340 bytes of audio encoded in base64 chunk and the avatar seems to be lagging while speaking.

I am not sure if this is the root cause. Can someone help on this?