Hacker News Reader

Comment by Cort3z | Hacker News Reader

I usually find Claude to be my favourite flavor of LLMs, but I still pay for ChatGPT because their voice offering is so great! I regularly use it as an "expert on the side" when I do other things, like doing bike repairs. I ask it things like "how do I find the min/max adjustments on my particular flavor of front derailleur", or when cooking, and my hands are dirty, I can ask stuff like "how much X do I usually need for Y people", and so on. The hands-off feature is so great when my hands are literally busy doing some other thing.

I really wish Claude had something similar.

View on HN

ChatGPT advanced voice mode really is surprisingly excellent - I just wish it:

1) would give you more time to pause when you’re talking before it immediately launches into an answer

2) would actually try to say the symbols in code blocks verbatim - it’s basically useless for looking up anything to do with code, because it will omit parts of the answer from its speech.

Yeah I have to manually hold it down every time I talk. I have a lot of pauses and simply would not be able to interface with that without that option. It’s why I essentially can’t use Gemini voice mode

I think voice interface is the real killer app of LLMs. And the advance voice mode was exactly what I was waiting for. The pause between words issue is still a problem though, I think being able to just hit enter when done would work best.

Pro tip; if you’re preparing for a big meeting eg an interview, tell ChatGPT to play the part of an evil interviewer. Give it your CV and the job description etc. ask it to find the hardest questions it can. Ask it to coach you and review your answers afterwards, give ideal answers etc

after a couple of hours grilling the real interview will seem like a doddle.

> There's less usage of voice mode on the enterprise and power users side but that will happen eventually. - Anthropic CEO 21 jan. [0]

[0] https://youtu.be/snkOMOjiVOk 01:30

Is it possible to use ChatGPT voice feature in a similar manner to Alexa where I only need to say an activation word? I’m aiming to set up a system for my 7-year-old son to let him engage in conversations with ChatGPT as he does with Alexa.

I assume it would be possible to make yourself with the OpenAI api together with a locally run voice model to only detect the activation word. There might be of the shelf solutions for this, but I am not aware of any.

Reply on news.ycombinator.com