OpenAI has started rolling out the long-awaited Advanced Voice Mode for paid Pro users.
This format allows for more natural communication with LLM, resembling a regular conversation.
P.S. I have to admit that I tried using the voice feature of ChatGPT for dictating thoughts for a while, aiming to get a summary by the end of the day and transfer it to my journal - but the idea turned out to be unworkable. The voice feature loses track in a few minutes, throws an error, and ultimately the whole experiment is flawed. I’ll give Advanced Voice Mode a second chance, but it seems I won’t escape the need to write a bot in Telegram that will accept voice messages and perform agent tasks based on them.