Most active commenters

    ←back to thread

    684 points prettyblocks | 17 comments | | HN request time: 0.932s | source | bottom

    I mean anything in the 0.5B-3B range that's available on Ollama (for example). Have you built any cool tooling that uses these models as part of your work flow?
    1. flippyhead ◴[] No.42785739[source]
    I have a tiny device that listens to conversations between two people or more and constantly tries to declare a "winner"
    replies(14): >>42785781 #>>42785791 #>>42785949 #>>42785970 #>>42785979 #>>42786455 #>>42786672 #>>42787108 #>>42788174 #>>42788937 #>>42789840 #>>42791711 #>>42807514 #>>42890452 #
    2. pseudosavant ◴[] No.42785781[source]
    I'd love to hear more about the hardware behind this project. I've had concepts for tech requiring a mic on me at all times for various reasons. Always tricky to have enough power in a reasonable DIY form factor.
    3. oa335 ◴[] No.42785791[source]
    This made me actually laugh out loud. Can you share more details on hardware and models used?
    4. econ ◴[] No.42785949[source]
    This is a product I want
    5. amelius ◴[] No.42785970[source]
    You can use the model to generate winning speeches also.
    6. jjcm ◴[] No.42785979[source]
    Are you raising a funding round? I'm bought in. This is hilarious.
    7. hn8726 ◴[] No.42786455[source]
    What approach/stack would you recommend for listening to an ongoing conversation, transcribing it and passing through llm? I had some use cases in mind but I'm not very familiar with AI frameworks and tools
    8. eddd-ddde ◴[] No.42786672[source]
    I love that there's not even a vague idea of the winner "metric" in your explanation. Like it's just, _the_ winner.
    9. mkaic ◴[] No.42787108[source]
    This reminds me of the antics of streamer DougDoug, who often uses LLM APIs to live-summarize, analyze, or interact with his (often multi-thousand-strong) Twitch chat. Most recently I saw him do a GeoGuessr stream where he had ChatGPT assume the role of a detective who must comb through the thousands of chat messages for clues about where the chat thinks the location is, then synthesizes the clamor into a final guess. Aside from constantly being trolled by people spamming nothing but "Kyoto, Japan" in chat, it occasionaly demonstrated a pretty effective incarnation of "the wisdom of the crowd" and was strikingly accurate at times.
    10. nejsjsjsbsb ◴[] No.42788174[source]
    All computation on device?
    11. prakashn27 ◴[] No.42788937[source]
    wifey always wins. ;)
    12. deivid ◴[] No.42789840[source]
    what model do you use for speech to text?
    13. TechDebtDevin ◴[] No.42791711[source]
    Your SO must really love that lmao
    14. econ ◴[] No.42807514[source]
    Tell me it also does sports style commentary on the ongoing debate. My mental image requires it.
    15. flippyhead ◴[] No.42890452[source]
    Heh, I made this comment and forgot to check back -- I'm always missing stuff on HN because of this!

    If anyone is still paying attention, email me at hi@seikai.tv and I'll see if I can send you one.

    replies(2): >>42981978 #>>42992234 #
    16. Shonku_ ◴[] No.42981978[source]
    Yeah I'm still paying attention!
    17. ultrasounder ◴[] No.42992234[source]
    Sounds cool! In fact, this can be applied to other areas such as "debate monitoring" for debate competitions