Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GPT‑Live

openai.com

Post reply on HN
Print view View original

361–370 of 556 posts

Re: GPT‑Live

#361

Post by bariswheel »

athyuttamre wrote: post #334
WhitneyLand wrote: post #224

Much better than it was before but it’s still significantly weaker than a direct chat. For example I asked “Why should LLM attention use dot product instead of cosine similarity, being that we often hear vector magnitude does not encode most of the useful information needed”? The voice response was directionally right but lacked detail and was a little hand wavy. The answer to the same question in a text chat was muc…

A lot of this is steerable! Our current personality is optimized for brainstorming and conversations, but you can provide custom instructions to ask it to go deep and give you info-dense or more technical answers.

I found myself probing more and more questions to have it do that, I wish it could remember to provide details answers, I thought the replies were way too short and hand-wavy like one of the prior comments suggested.

Re: GPT‑Live

#362

Post by bariswheel »

I don't think the voice feature should act like a human, but 'complement' it. I already have friends I can talk to. Perhaps for quick answers it might be helpful but Google already does that for me and I don't have to worry about having to 'archive' the chat later and creates clutter.

I really hope at some point we can 'option click' or whatever, and choose multiple threads to archive. It takes FOREVER to archive the cluttery chats one by one. That's my one wish feature, please make batch cleanup of my client reasonably easy. Imagine having to delete one file at a time in a folder of 50 plus files. Barf.

Re: GPT‑Live

#363

Post by overgard »

throwuxiytayq wrote: post #110
jonstaab wrote: post #76

This is the opposite direction AI should be going. Human relationships are the most valuable thing we have, and so, naturally, technology seeks to intermediate and now replace them. I'm not Catholic, but this podcast presents a very interesting argument against talking to AI as if they were human: https://newpolity.com/podcasts-hub/debate-chatbots

How would talking to an AI as if it were not human sound? You can probably set your system prompt to insert “beep boop” between sentences and make it refer to itself as “Cybertron9000 Personal Computing Device” if that’s what you like. Is that an improvement? Or are you against voice computer interfaces altogether?

I think you could make it impersonal sounding. Like, right now ChatGPT throws in a lot of cheeky things. They don't really bother me because I don't anthropomorphize it, but, a lot of otherwise smart people are struggling with anthropomorphizing these things right now (Richard Dawkins..) so it strikes me that the less personality they have the better. (If you want personality just throw it into the prompt, ya know?) To their credit I think this is customizable in ChatGPT right now (it's been a while since I looked), but "impersonal" should probably be the default until society adjusts to these things.

Re: GPT‑Live

#364

Post by wat10000 »

jjulius wrote: post #150
simonw wrote: post #19

I had preview access to this one for a few weeks. It's very good. I had one conversation that lasted a full hour while I was walking the dog, got some good brainstorming done against one of my projects. The best feature is that it can delegate questions out to GPT-5.5 in the background, so you're no longer restricted to a voice model that's several years behind the frontier. I did report a fun bug with it though: it…

I highly recommend simply enjoying the walk.

Sometimes I want to learn something while I walk. Sometimes I want to listen to stories. Sometimes I want to just be present. All are good. I highly recommend letting people enjoy whatever they enjoy.

Re: GPT‑Live

#365

Post by supern0va »

jjulius wrote: post #150
simonw wrote: post #19

I had preview access to this one for a few weeks. It's very good. I had one conversation that lasted a full hour while I was walking the dog, got some good brainstorming done against one of my projects. The best feature is that it can delegate questions out to GPT-5.5 in the background, so you're no longer restricted to a voice model that's several years behind the frontier. I did report a fun bug with it though: it…

I highly recommend simply enjoying the walk.

Every time something along these lines is posted, comments like this show up.

The thing I don't get is...no one would say this about listening to a podcast or audiobook on a walk.

I'm not sure why people choose to demonize this specific use of time during walks.

Re: GPT‑Live

#366

Post by SmirkingRevenge »

throwuxiytayq wrote: post #309
SmirkingRevenge wrote: post #244

Earlier quoted context omitted.

My preference would be to turn down the fake emotional expressions and notes in the responses. No cheeky quips, etc. Otherwise it's kind of like being manipulated by a psychopath

I would also like to tone those down, and prefer if they just focused on creating a nice, clear, natural-sounding voice. I don’t think the “emotional roleplaying” adds anything to what I perceive as a computer interaction. But that’s precisely the opposite of the problem the other commenters seem to be voicing; I have no problem recognizing what I’m interacting with, and I don’t worry about being manipulated - not by…

Maybe it's just that conversing with something or someone that is acting manipulative can be unpleasant in and of itself, even if the manipulation is not successful.

Each time I hear the bot emote, it's like a subtle suggestion to just play along and anthropomorphize it - it feels weird, silly and annoying

Re: GPT‑Live

#367

Post by squidsoup »

amarant wrote: post #245
tough wrote: post #212

Earlier quoted context omitted.

the cybertruck seems designed for this KITT fantasy of yours

Dammit! I live in Europe and the cyber truck isn't even available here.. That said a huge pickup truck is about as far as you can get from a Camaro... Then again I'm not exactly David Hasslehoff myself either... Meh if it talks that's close enough!

> Dammit! I live in Europe and the cyber truck isn't even available here..

Europeans are notorious for their good taste.

Re: GPT‑Live

#368

Post by pseudosavant »

100ms wrote: post #319
100ms wrote: post #146

Earlier quoted context omitted.

Star Trek computer voice model is something I have yet to encounter, and I've looked repeatedly :) It's not about a specific voice, it's the fact they managed to capture "I am a utility" perfectly in the voice. Our modern friends do not want to be thought of as a utility, but to engender trust and agency all of their own and that's a huge problem for me.

I thought to try voice cloning with dots.tts ( https://huggingface.co/spaces/rednote-hilab/dots.tts ), the result is pretty good, but likely wouldn't be fast enough to use on a quasi-realtime basis: Input clip: https://vocaroo.com/19QtEPtwTjOS Prompt text: There are 14 varieties of tomato soup available from this replicator. With rice, with vegetables, Bolian style, with pasta specify hot or chilled. Output: https://…

That is pretty incredible with such a small amount of input. Makes me want to check out dots.tts myself!

Re: GPT‑Live

#369

Post by himata4113 »

motoboi wrote: post #315

Oh, god. They are marketing it as an old people artificial friend. Probably will blame users in the future when they get attached and have ai-induced psychosis. Disgraceful. On the technological side, it's a marvel!

I think this is fine as long as the models stay honest. I.e. refuse to be a friend, girlfriend, partner or pretend to be someone.
Post reply on HN