How do I get the called party's audio stream while playing audio

[fish_cat]

hello everyone

Hello, I want to stream both the parties audio separately to a web socket for real time transcription and diarization(speaker labelling). I am able to record the audio separately using monitor for both agent and costumer but i want to steam the audio

Sure, you can. You can use two snoop channels and put them into the bridge with externalMedia. Snoop channel receives only one client audio. See here for example or here

I saw this post, and then I adopted the second asterisk-ari/vosk_ari.py scheme.
I want to play a greeting audio for the other party when the phone is connected, the audio is 15 seconds long, and then monitor the other party’s speech during the 15 seconds and transcribe it in real time.
I tried to add a line to the init method: channel.play(media=sound:custom/welcome)

Result: During the first 15 seconds of audio play, there is no real-time transcription, it is transcribed at the end of the audio play, and it will also transcribe the speech during the audio play,
I guess the user’s speech, during the first 15 seconds of audio play, is added to the queue or cache, or other places, how can I modify it, can you help me