hi team, i have build an agent in th livekit that it very slow iam using deepgram nova-3 as stt, gpt4.1-mini as LLM and cartesia TTS. still there is some delay in the response like probably 1-2 seconds. what could be the reason? can someone suggest the best stt and tts

Many times following this can help: https://kb.livekit.io/articles/4490830410-how-can-i-reduce-latency-in-voice-pipeline-agents-using-stt-tts-and-llm

I've found OpenAI LLM to be slow and unrealiable (often timing out) when used with a pipeline agent. RealTime is fine. It suprised me tbh - but as soon as I switched LLM provider, the problems went away.

which llm provider that you switched?

groq or google seem to be more reliable - google seems better for llm quality

iam using multiple tool calling with the llm. what is the best llm do you suggest?

the only problem I've experienced with tool-calling is with Google REALTIME - but that's another story.