How do I send the STT results and the LLM output to the client

[yeyupiaoling]

• pipecat-ai 0.0.42.dev63
• Ubuntu 22.04
• Client：Android

I wanted to be able to display a transcript of the conversation on the client, so I needed pipecat-ai to be able to send STT results and LLM output to the client.

What should I set?

My code snippet is as follows:

async with aiohttp.ClientSession() as session:
    transport = DailyTransport(
        room_url,
        token,
        bot_name="Bot",
        params=DailyParams(
            audio_out_enabled=True,
            transcription_enabled=False,
            vad_enabled=True,
            vad_analyzer=SileroVADAnalyzer(params=VADParams(stop_secs=0.2)),
            vad_audio_passthrough=True,
        ),
    )
    stt = AzureSTTService(name="STT", **configs.azure_stt)
    tts = AzureTTSService(name="TTS", **configs.azure_tts)
    llm = OpenAILLMService(name="LLM", **configs.llm)

    messages = [{"role": "system", 'content': get_system_prompt()}]

    tma_in = LLMUserResponseAggregator(messages)
    tma_out = LLMAssistantResponseAggregator(messages)

    pipeline = Pipeline(
        [
            transport.input(),  # Transport user input
            stt,  # Speech-to-text
            tma_in,  # User responses
            llm,  # LLM
            tts,  # TTS
            transport.output(),  # Transport bot output
            tma_out,  # Assistant spoken responses
        ]
    )

    task = PipelineTask(
        pipeline,
        PipelineParams(
            allow_interruptions=True,
            enable_metrics=False,
            report_only_initial_ttfb=False,
        ),
    )