Is it possible to make the transcription always transcribe in english?

Is it possible to configure the transcription to always be in English, regardless of the spoken language?

I’m experimenting with the application and trying to create a real-time translation agent. By “real-time,” I mean truly instant translation.

As soon as someone starts speaking, I need to receive the translation of what they are saying immediately in English

Hi @wignesh and welcome to the community!

I’ve seen a colleague of mine do it on a Macbook as a real-time Zoom translator using the ChatGPT desktop app and Amoeba Loopback.

Steps:

  1. Open ChatGPT desktop app
  2. Loopback Audio Driver in Pass-Through mode to capture the output of Zoom as input/micro for chatGPT
  3. Set the microphone input of Zoom and your audio settings to “Loopback device”
  4. Prompt in ChatGPT: “Please act as a real-time translator for me and translate what you hear into English.”

Thanks for sharing this. My goal is to receive the user’s spoken words in English. However, the current transcription from the Realtime API is returning the spoken language as-is. I’ve tried using prompt instructions, but it doesn’t seem to be working.