Openai API voice for twilio

Community
1

I would like to inquire whether it is possible to leverage the GPT-4 voice capabilities through the OpenAI API to enable a Twilio-powered agent to make client calls with a voice quality that matches the natural and fluid sound of GPT-4’s voice. Specifically, my question revolves around achieving this level of natural voice quality for phone interactions.

I have attempted to integrate an ngrok server, Twilio, OpenAI API, and GoHighLevel to automate actions such as appointment bookings. However, despite my efforts, I was unable to replicate the high-quality, natural voice of GPT-4 using the OpenAI API’s text-to-speech functionality. Could you provide guidance on how to achieve this, or recommend alternative approaches to obtain a similarly natural-sounding voice?

2

Voice modality on gpt-4o is currently not available on the API.

Best option is to use TTS models to generate voice, as of writing this post.

1 Like
3

So voice model and tts is not the same

4

Voice mode is used on ChatGPT app.

TTS (Text To Speech) models can be used with the Speech endpoint in the Audio API

1 Like
5

Okay, I’ll rephrase my question because I didn’t quite get the answer I was looking for. Is text-to-speech (TTS) the same as GPT’s voice in terms of sound and fluidity? That’s what I’m trying to clarify.

Thanks for your help!