agents icon indicating copy to clipboard operation
agents copied to clipboard

Build real-time multimodal AI applications 🤖🎙️📹

Results 238 agents issues
Sort by recently updated
recently updated
newest added

Currently VoiceAssistant synchronizes transcription with audio playout. However, some would like the full transcription to appear immediately. we should make this an option

**Error description** After the room ends normally, the agent reports an error, **Env** livekit-server: 1.6.1 agent:0.8.10 **To Reproduce** 1. lk room join --identity test-lijiliang room-lijiliang, then the agent will automatic...

`before_llm_cb` is only called when there is an audio message from the user If there is a text message, `before_llm_cb` is not called. It seems like, for consistency purposes, this...

Hi! Is it correct that the livekit-azure-plugin only works with the Azure native STT and TSS services and that AzureOpenAI's STT and TSS are note in the livekit SDK yet?...

This PR reuses the `SpeechHandle` ID to serve as an _inference ID_, enabling LLM invocations to be joined against text in the chat log. This allows consumers to commit side...

### How to reproduce Not able to reproduce consistently, but it happens occasionally with the example code provided in https://github.com/livekit/agents/blob/main/examples/voice-assistant/minimal_assistant.py Might be able to trigger it locally if you can...

- Added comprehensive docstring to the `WorkerOptions` class explaining the purpose and usage of each argument. - Included descriptions for key attributes such as entrypoint function, request handler, load function,...

Wondering if new support for [Structured Outputs](https://openai.com/index/introducing-structured-outputs-in-the-api/) can be added for openai models?

Now consumers can log all LLM completion requests like this: ```py def will_log_completion_event(cht_ctx, collected_text, tool_calls, interrupted): my_logger(cht_ctx, collected_text, tool_calls, interrupted) assistant = VoiceAssistant( vad=silero.VAD.load(), stt=deepgram.STT(), llm=openai.LLM(), tts=openai.TTS() chat_ctx=initial_ctx, will_log_completion_event=will_log_completion_event )...

I noticed that PRs require a CLA. Personally this greatly reduces the chances that I would use this project. A CLA basically says to me that the project will change...