My Agents are choppy this week. For example, if ...
# agents
b
My Agents are choppy this week. For example, if a user says: "I am having trouble with this thing..." then sometimes the agent says "I understand how that can be frustrating" and other times the agent skips "I", and just says "understand how that can be frustrating". When it happens it has a glitchy vibe where you can tell the "I" got skipped or cut off. Other times an agent will start a sentence, then stop, then start a new sentence. I was curious if could be related to the VAD settings, but no luck there so far. Claude said to use
session.say()
instead of
session.generate_reply(tool_choice="none")
and that did clean up the initial greetings, but probably because they're hard coded now. For context this is with the restaurant starter recipe that was converted to use inference. This issue happens locally without SIP, and when calling livekit cloud sip / telnyx. I am just wondering where to optimize or fix here. Could it be related to using inference, or maybe VAD, or is it the actual STT and TTS models? I was also curious if it could be because I moved my prompts over to separate yaml files. Claude code was saying that's not the culprit. Anyone else having similar issues? If so any luck solving?