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

Introducing ChatGPT and Whisper APIs

openai.com

Post reply on HN
Print view View original

631–640 of 696 posts

Re: Introducing ChatGPT and Whisper APIs

#631

Post by WolfOliver »

minimaxir wrote: post #2

> It is priced at $0.002 per 1k tokens, which is 10x cheaper than our existing GPT-3.5 models. This is a massive, massive deal. For context, the reason GPT-3 apps took off over the past few months before ChatGPT went viral is because a) text-davinci-003 was released and was a significant performance increase and b) the cost was cut from $0.06/1k tokens to $0.02/1k tokens, which made consumer applications feasible wit…

I think you will consume a lot of tokens very quickly as you have to send the entire chat history back and forth if you want to append another chat message.

Re: Introducing ChatGPT and Whisper APIs

#632

Post by wriggler »

graderjs wrote: post #615
wriggler wrote: post #608

Earlier quoted context omitted.

Hey, nice app. Can you add support for non-English transcription/translation, though?

Thanks! So, OK, if you write a review and provide your honest feedback about this on the App store, I will definitely consider doing that! Sound like a bad idea? :)

Done

Re: Introducing ChatGPT and Whisper APIs

#633

Post by graderjs »

wriggler wrote: post #632
graderjs wrote: post #615

Earlier quoted context omitted.

Thanks! So, OK, if you write a review and provide your honest feedback about this on the App store, I will definitely consider doing that! Sound like a bad idea? :)

Done

Thank you sir! I will look for it. Stay tuned for updates! I might just consider putting that in soonish :) But I might not. I don't know. Can't guarantee anything about it right now. Thank you! asking and telling me about this

Re: Introducing ChatGPT and Whisper APIs

#634

Post by graderjs »

qwertox wrote: post #624
graderjs wrote: post #597

Earlier quoted context omitted.

Transcript of 20230302 170602.m4a (at 2023-03-02 1711.02).txt Hey friend, your use case sounds really interesting. Actually that's why I created this app initially. I really love riding around on my bike in the city and doing voice memo debriefs about whatever. I also like to do it walking around outside. And as you say, the trouble with that is wind distortion. Full stop. On a day where it's not too windy, it's not…

I'm not preprocessing the recording. The thing I mentioned with considering VAD, if that's what you're referring to, is that I should somehow trim the files. In order to minimize my interaction with my phone during the bike ride, I press a button which records 1 minute of audio. If I know that I need more time, I press it again before the minute ends, this then starts a second recording in parallel which also lasts o…

This is great! Thank you. We’re very similar actually: I also tried getting chat GPT to correct transcripts for errors but when I tried like 3 weeks ago it couldn’t manage. I just use voice memos app and let it run. I just talk and don’t think about the file. So I only have 1 file and my WisprNote app removes nearly all the non speech and passes it to whisper. I think there’s actually a voice memo setting on MacOS that will cut silence automatically but I don’t use it.

Re: Introducing ChatGPT and Whisper APIs

#635

Post by addandsubtract »

ghshephard wrote: post #365
evilduck wrote: post #323

Earlier quoted context omitted.

Thanks, what about broader news (GPT, Bing, Llama, etc)? The Stable Diffusion sub is only image AI oriented.

Twitter. Folllow your top 10 or so ML/AI news summarizers. There is enough new information every day to keep you busy reading new papers, APIs, technologies. Honestly the "This happened in the last week" is more information than anybody can fully wrap their heads around, so you just have to surf the headlines and dig into the few things that interest you.

Can you recommend some accounts to follow?

Re: Introducing ChatGPT and Whisper APIs

#636

Post by osigurdson »

cubefox wrote: post #549
ShamelessC wrote: post #352

Earlier quoted context omitted.

It’s stateful in the web demo. But they do so by prepending chat history to new requests and automatically summarizing history once the model’s context window is exceeded.

I don't think they are doing summaries actually. Just cutting off the oldest parts of the conversation.

Summarizing would make sense however.

Re: Introducing ChatGPT and Whisper APIs

#637

Post by tomhallett »

Has OpenAI said anything (in their TOS/etc) about who owns the copyright/ip on the code that it generates?

I know it’s a legal minefield of a question, just curious if they have said “we won’t sue you if you copy/paste this code in your app” publically or anything.

Re: Introducing ChatGPT and Whisper APIs

#638

Post by dnadler »

WolfOliver wrote: post #631
minimaxir wrote: post #2

> It is priced at $0.002 per 1k tokens, which is 10x cheaper than our existing GPT-3.5 models. This is a massive, massive deal. For context, the reason GPT-3 apps took off over the past few months before ChatGPT went viral is because a) text-davinci-003 was released and was a significant performance increase and b) the cost was cut from $0.06/1k tokens to $0.02/1k tokens, which made consumer applications feasible wit…

I think you will consume a lot of tokens very quickly as you have to send the entire chat history back and forth if you want to append another chat message.

This is exactly right. It's cheap, but not as cheap as people think. Conversations get exponentially more expensive with message length.

Re: Introducing ChatGPT and Whisper APIs

#639

Post by huqedato »

ar9av wrote: post #539
minimaxir wrote: post #2

> It is priced at $0.002 per 1k tokens, which is 10x cheaper than our existing GPT-3.5 models. This is a massive, massive deal. For context, the reason GPT-3 apps took off over the past few months before ChatGPT went viral is because a) text-davinci-003 was released and was a significant performance increase and b) the cost was cut from $0.06/1k tokens to $0.02/1k tokens, which made consumer applications feasible wit…

Pricing of this model seems less per token level but you have to send the entire conversation each time, and the tokens you will be billed for include both those you send and the API's response (which you are likely to append to the conversation and send back to them, getting billed again and again as the conversation progresses). By the time you've hit the 4K token limit of this API, there will have been a bunch of…

You're right. And this is critical for large text (summarization, complex prompting etc.). Thats's why I'll continue to use text-davinci-xxx for my project.

Re: Introducing ChatGPT and Whisper APIs

#640

Post by dragonwriter »

visarga wrote: post #535
dragonwriter wrote: post #451

Earlier quoted context omitted.

> The only way to a post-scarcity economy There is no way to a post-scarcity economy, the phrase is a contradiction, and scarcity is an inescapable consequence of human desire.

There's no scarcity of information and media.

> There's no scarcity of information and media.

There is scarcity of information and media of the kind people find valuable. If there is a sense in which the broader statement is true, it is irrelevant for that reason.

Post reply on HN