Integrating openai's audio gpt-4o-audio-preview model with agents

    @step
    async def handle_audio(self, ctx: Context, ev: AudioInput) -> ProcessedInput:
        # Process audio input
        result = self.process_audio(ev.audio_path)
        return ProcessedInput(result=result)

...
    # use this function to process audio 
    def process_audio(self, audio_path: str) -> str:
        """Process audio input and return a description."""
        messages = [
            ChatMessage(
                role="user",
                blocks=[
                    AudioBlock(path=audio_path, format="wav"),
                    TextBlock(text="Describe the content of this audio."),
                ],
            )
        ]
        llm=self.agents[self.root_agent].llm 
        response =  llm.chat(messages)
        return str(response)