kshmir
kshmir on Hacker News
Much safer to just use Meta Business Agent, avoids bans.
The big difference is that we support attachments and transcribe the audios, we've been in business for over 4 years.
Interesting approach, we did kaptionai.com/mcp with a slightly different approach, the MCP runs on your own WhatsApp Web or in our own app, it can run locally or through our own bridge (which is open source as well).
Except you probably pay cents in tokens for that, or included with your sub.
how do I follow you? I have a small 5090 doing inference all the time and I barely use tts but a lot of asr, mostly whisper, I ported your tech report for tts and implemented some improvements on my whisper inference based on your tech report as well!
would love to talk sometime!
Thought there were only 2 options!
Thanks for your article I set it in my 5090 server and took some ideas to improve the whisper inference I also have.
Running asr+llm+tts now :)
Besides the numbers being stupidly overblown, this post shows why Europe is in a unstoppable death spiral.
Bootlickers gonna bootlick… aren’t you part of the problem?
It's never about shooting.
Says someone who never owned a Tesla.
I still don't care. I may just buy another Tesla and a handgun :)
I just love the Tesla, why would I care about what other people say? It works better everyday.
Why pay what you're asking instead of using dragonfly or something like that and just putting a beefier node?
OP, contact me, also from Argentina.
Please don’t say “open” when it is not.
Yeah things are absolutely fine for kids today, keep it up champ.
Lemme guess your state.
What is this trained on? It doesn't seem to look on the web for my product at all... doesn't quite the reviews...
Are you just using an LLM? I asked to write code, and it did...
We probably need superconductors for that.
> mandastes to use self-driving cars
Lol you’re insane. I have a self driving car I own
What happens when Metabase releases this? (Asking without malice!)
We think them releasing those features will increase our exposure, but most computers won’t run models and most phones won’t stream AI.
Also, Meta won’t run those models in the cloud for privacy reasons (being, most people don’t trust them).
We use a mix of distilled whispers on custom hardware to make ends meet.
We’re also planning on releasing more customer (and business) oriented features, but not based on AI.
We just made the whatsapp web we wanted and a price tag on it :)
Posted by
kshmir on Wednesday August 07, 2024 @09:50PM
from the show-hn dept.
https://kaptionai.com
Show HN: KaptionAI - Automate Transcription and Summarization for WhatsApp Voice MessagesHey HN,We're excited to introduce KaptionAI, a tool we’ve developed to streamline WhatsApp communication by using AI to transcribe and summarize voice messages.Key Features:• Voice to Text: Converts WhatsApp audio messages to text quickly and accurately.
• Summarization: Generates concise summaries of long...
You’ll need gpu acceleration
Just set quotas by default…