Hi,
currently I want to use a text-in, audio-out model to build my chatbot. So for the WebRTC widget, I think I only need to use the mode= "recieve" when init it.
audio_text = WebRTC(
mode="receive",
modality="audio",
)
audio_text .stream(
fn=my_stream_handler
inputs=[input_text_box],
outputs=[audio_text ],
time_limit=600,
concurrency_limit=WEB_RTC_CONCURRENCY,
)
So for my customized stream handler implementation, how should I take care of the implement of the ".receive" method?
I think since I am using recieve only mode, it wont get any audio frame as input right?