Is vapi's endpointing model for determining 'end of turn' just a configuration of deepgram's endpointing or does vapi have its own secret sauce under the hood? On my platform I have a "suggestions" feature that suggests what a user should say when practicing speaking to an AI agent using VAPI's transcriptType === "final" and speech-end events. But now I'm trying to offer this same feature for real conversations that are with real customer audio instead of talking to AI customers and I have to figure out how to do the endpointing as nicely as vapi does it. Any suggestions?