instead of using function calling, it is not reliable always. What else can I use

Thanks for reaching out, Kruthik. I'm an AI assistant and I'm here to help. Give me a moment while I research this topic.

for livekit 1.0

In LiveKit 1.0, the function calling system has been replaced with a new approach called "tool definition and use." Here are the key updates: • The

@llm.ai_callable

decorator has been replaced with the

@function_tool

decorator. • Functions defined with

@function_tool

are automatically accessible to the LLM, eliminating the need for a separate

llm.FunctionContext

class. • Argument types are inferred from the function signature and docstring, and annotated types are no longer supported. • Functions now take a

RunContext

object, which provides access to the current agent state. This new system simplifies the process of defining and maintaining tools for agents. Let me know if you need further clarification! Sources: Agents 0.x migration guide | LiveKit Docs

is there any other better way to implement functions instead of using function_tool

In LiveKit 1.0, the

@function_tool

decorator is the standard and recommended way to define tools for agents. Currently, there are no direct alternatives to this method in version 1.0, as it has replaced the older

@llm.ai_callable

decorator from previous versions. This approach simplifies tool definition by making them automatically accessible to the LLM without requiring additional setup. If you have specific requirements or challenges with

@function_tool

, feel free to share, and I can assist further! Sources: Function calling with Voice Agents | LiveKit Docs | Agents 0.x migration guide | LiveKit Docs