331–340 of 1001 posts
Few people are talking about it but... what do you think about the very over-the-top enthusiasm? To me, it sounds like TikTok TTS, it's a bit uncomfortable to listen to. I've been working with TTS models and they can produce much more natural sounding language, so it is clearly a stylistic choice. So what do you think?
Big questions are (1) when is this going to be rolled out to paid users? (2) what is the remaining benefit of being a paid user if this is rolled out to free users? (3) Biggest concern is will this degrade the paid experience since GPT-4 interactions are already rate limited. Does OpenAI have the hardware to handle this? Edit: according to @gdb this is coming in "weeks" https://twitter.com/gdb/status/1790074041614717…
This might mean GPT-5 is coming soon and it will only be available to paid users.
From my casual conversations, not that many people are paying for GPT4 or know why they should. Every conversation even in enthusiast forums like this one has to be interjected with "wait, are you using GPT4? because GPT3.5 the free one is pretty nerfed"
just nuking that friction from orbit and expanding the GPT store volume could be a positive for them
If your wallet is large enough, you can make 2 GPTs sing just as easily as you can make 100 GPTs sing.
What can you do with a billion GPTs?
I am observing an extremely high rate of text hallucinations with gpt-4o (gpt-4o-2024-05-13) as tested via the API. I advise extreme caution with it. In contrast, I see no such concern with gpt-4-turbo-preview (gpt-4-0125-preview).
If true, makes me wonder what kind of regression testing OpenAI does for these models. It can’t be easy to write a unit test for hallucinations.
My brother who can't see correctly, will use this to cook a meal without me explaining this to him it's so cool.
People all around the world will now get real-time AI assistance for a ton of queries.
Heck - I have a meeting bot API company (https://aimeetingbot.com) and that makes me really hyped!
Kicking off another training wave is easy, if you can afford the electricity, but without new, non-AI tainted datasets or new methods, what’s the point?
So, in the meantime, make magic with the tool you already have, without freaking out the politicians or the public.
Wise approach.
In the video where the 2 AI's sing together, it starts to get really cringey and weird to the point where it literally sounds like it's being faked by 2 voice actors off-screen with literal guns to their heads trying not to cry, did anyone else get that impression? The tonal talking was impressive, but man that part was like, is someone being tortured or forced against their will?
I think this demo is more for showing the limit like "It can sing isn't it amazing?" than being practical, and I think it perfectly served the purpose.
I agree that the tortured impression. It partly comes from the facial expression of the presenter. She's clearly enjoying pushing it to the edge.