use OpenAI's TTS in place of the SpeechSynthesisUtterance API

gregsadetsky / sagittarius

A GPT-4/Gemini Voice/Video Exploration Tool

http://sagittarius.greg.technology/

685 stars 92 forks source link

use OpenAI's TTS in place of the SpeechSynthesisUtterance API #5

Open wobsoriano opened 9 months ago

wobsoriano commented 9 months ago

This PR integrates OpenAI's TTS in place of the browser's SpeechSynthesisUtterance API.

Pros

Audio API has better browser support.
Natural-sounding audio.

Cons

Longer response times due to the additional fetch call.
Additional cost

gregsadetsky commented 9 months ago

hey, thank you for this! I think that perhaps we could have it as an option so that people could still use the browser built-in TTS if they wanted, but this option for better voices would be available as well.

I'm definitely concerned with adding any additional delays (i.e. api calls), as anything to get interactions as close as possible to real time is much more important in my opinion

thanks again