dots-hyprland icon indicating copy to clipboard operation
dots-hyprland copied to clipboard

Published 20 hours ago • verified publisher icon end-4

[Feature] GPT-4o with audio support

Open codewithkenzo opened this issue 9 months ago • 3 comments

Mr. end-4, I know you want it too

codewithkenzo avatar May 15 '24 05:05 codewithkenzo

Can't wait for OpenAI to release GPT-4o with stt and tts support. :o

H0mire avatar May 16 '24 17:05 H0mire

oxygen api provides that for free idk if it's real but it does use emojis like the gpt4o on poe.com 4 or 4o? no clue

idk how to include sound yet

end-4 avatar May 17 '24 11:05 end-4

Yeah currently "Audio" is usually generated through a tts service, which you would have to integrate separately. OpenAI hinted that they will release the GPT 4o with Audio processing, which is basically native tts and stt without a separate model or service. This Results to a low latency like a normal human conversation and capability to process emotional expression.

H0mire avatar May 17 '24 13:05 H0mire