Hey everyone,
I’ve tested out Voice Assist in the web application - works great for answering general questions, but at the moment it’s not usable for e.g. coding or similar work related use cases, because it is only capable of answering with voice.
If you’d give the application a canvas to show things, it would be much more helpful. Example use-case: “My zshell is slow on startup - how to I find out, what is causing that?”
The ChatGPT Chat interface will give me helpful explanations along with code examples, that I can look over, understand and copy-and-paste - voice assist starts to read back the code to me, which is kind-of pointless (maybe helpful for visually impaired users, but not for the bulk of users).
Instead, ChatGPT Voice could show the answer (as in the chat interface) below the voice-bulb-circle-portrait (in lack of a better word). That way, we’d get the best of both worlds - have a nice conversation, but with the ability to copy-and-paste the provided answers.
By making it a “canvas” instead of a text-interface, you’d also be able to use other output media like images.
Thanks for evaluating this!