GPT-4, ChatGPT, and Whisper(speech-to-text) Plugins for TouchDesigner
Hey! In this tutorial we'll go over 2 new components I developed to run OpenAI's Whisper (speech-to-text) and GPT-4(includes multiple versions) within TouchDesigner. The components work without any setup needed, just add your OpenAI API key. I also show how to extend the functionality of the components so you can use these models in your installations.
Link to the Project File is on my Patreon: / blankensmithing
Link to Open AI's Website: https://platform.openai.com/signup
Update
The component from this video now works with the new GPT-4 model. You need to get on OpenAI's waitlist to get access: https://openai.com/waitlist/gpt-4-api
The component defaults to GPT-3.5-turbo, so it works out of the box. If you get off the waitlist you can change the model from the component parameters, which is now listed under the API key parameter on the TD component.
0:00 Overview
1:03 Shader Generation Example
4:08 Priming messages for ChatGPT | Becoming Refik Anadol
7:23 API Key setup
9:46 In Depth use of the Whisper and ChatGPT Components
12:41 Building the connection from Whisper to ChatGPT