agents

agents copied to clipboard

Update README grammar

1

Changes: "**This class** completely wraps ..." is referring to the class in the singular. That means we need to make these changes: 1. “abstract away” → “abstracts away” (to match...

ShayneP

Trying to wrap my head around multimodal agent and openai realtime API :) I want to steer the conversation by managing the system context and my intuition was that I...

mrdrprofuroboros

When testing, I notice that sometime the agent doesn't receive the task even when the load is under the threshold. I put the log and found that the websocket didn't...

Max-Thuc

examples: added trigger-phrase agent example

5

s-hamdananwar

When `allow_interruption=False`, we should ignore user's input

2

When the agent is speaking with `allow_interruption=False`, we should not be processing any user input, instead of queuing up another response (only to play it out later). That response will...

davidzhao

OpenAI timeouts are too low / aren't handled well

2

The LLM client is configured with a 5 second read timeout. If the client times out (which it does very often with a short timeout), the stream is not resumed....

jezell

Implement Manual VAD Commit via Button for Controlled Speech Processing

5

I've implemented a button in the client that is supposed to ensure VAD (Voice Activity Detection) doesn't immediately commit my conversation and send it to the server. Instead, it should...

ChrisFeldmeier

Useing RAG with openai multimodal agent

3

Is there a sample code or can you guide to pass additional context to llm, like in this pipeline agents example with new openai multimodal example? https://github.com/livekit/agents/blob/main/examples/voice-pipeline-agent/simple-rag/assistant.py

Test-isom

[Very Important for LiveKit] Request add general mechanism to customize plugins of VoiceAssistant ASAP

4

Livekit bring very good RTC to world with OpenSource or Cloud, Awesome! But Livekit Agent has one big problem: The Livekit' VoiceAssistant ' Pipeline are hardcoded as combining VAD+STT+LLM+TTS ,which...

taylorgwei

Aggressive transcript mode / text response only mode

2

I think a common use case is to toggle between voice and text mode (like in the ChatGPT app among others). If the goal is to create a multimodal framework...

willsmanley