Live data from Hacker News

Show HN: A real time AI video agent with under 1 second of latency

news.ycombinator.com

211–220 of 264 posts

Re: Show HN: A real time AI video agent with under 1 second of latency

#211

Earlier quoted context omitted.

I see it the other way around. I think our human-human interaction style will “leak” into the way we interact with humanoid AI agents. Movie-Her style.

Mine certainly has. I type to ChatGPT much more like a human than a search engine. It feels more natural for me as it's context aware than search engines ever were. I can ask follow up questions and ask for more details about a specific portion or ask for the analysis I just walked it through to get the results I want to apply to another data set. "Now dump those results into a markdown table for me please."

Yeah... was thinking about that the other day. Is it wierd to say please to an AI? i'll say please, but i'll never correct my spelling. Sometimes it's garbled because i missed a space and a couple key strokes but it always understands.

Re: Show HN: A real time AI video agent with under 1 second of latency

#212

Felt like talking to a person, I couldn't bring myself to treat it like a piece of code, that's how real it felt. I wanted to be polite and diplomatic, caught myself thinking about "how I look to this person". This brought me thinking of the conscious effort we put in when we talk with people and how sloppy and relaxed we can be when interacting with algorithms. For a little example, when searching Google I default t…

I see it the other way around. I think our human-human interaction style will “leak” into the way we interact with humanoid AI agents. Movie-Her style.

Only if the AI gets annoyed when you don't treat it with respect.

Re: Show HN: A real time AI video agent with under 1 second of latency

#213
So... What's the new turing test? A test that stood for 50+ years is going to be completely ignored as a false test/ doesn't really mean anything? Because the turing test was text based, and this video based seems a couple of years from passing even a video based turing test.

Re: Show HN: A real time AI video agent with under 1 second of latency

#214

A question that's going to become very real very soon is this: If I video call someone and need them to prove they are human. What do I do? Initially it will be as easy as asking them to stand up and turn around, or describe the headlines from this morning's news. But that won't last long. What's the last thing an AI avatar will be able to, that any real human can do?

Meet in person :)

Re: Show HN: A real time AI video agent with under 1 second of latency

#215
post #132
post #102

Earlier quoted context omitted.

This looks awesome. Didn’t seem to hear me, but the video looks great. Can you share what models you are using? You say these are all open models.

The model doing the heavy lifting is https://github.com/Rudrabha/Wav2Lip Mic permissions on mobile are tricky, which might have been your issue? Note in this prototype you also need to hold the blue button down to speak.

Interesting. I didn’t think you could get anything close to realtime with Wav2Lip.

Re: Show HN: A real time AI video agent with under 1 second of latency

#216

I talked to your twin did you store my private info (face, voice)?

nope- we dont store any video/audio recordings of the sessions.

You'd have to enable that and similar to zoom, it would show on the screen that that is being recorded

Re: Show HN: A real time AI video agent with under 1 second of latency

#219

Functionality for a demo launch: 9.5/10 Creepiness: 10/10

I was just about to try it, but the idea of allowing Firefox access to my audio/video to talk to a machine-generated person gave me such a bad feeling, I couldn't go through with it even fuelled by my morbid curiosity.

I did it with my finger over the camera and it even commented on me having my finger over the camera!

Re: Show HN: A real time AI video agent with under 1 second of latency

#220
Good job on the launch and the write up. I'll be interested to play with this api.

I'm glad to see the ttft talked about here. As someone who's been deep in the AI and generative AI trenches, I think latency is going to be the real bottleneck for a bunch of use cases. 1900 tps is impressive, but if it's taking 3-5 seconds to ttft, there's a whole lot you just can't use it for.

It seems intuitive to me that once we've hit human-level tokens per second in a given modality, latency should be the target of our focus in throughput metrics. Your sub-1 second achievement is a big deal in that context.

Post reply on HN