104 karma · joined December 14, 2014
Source available API for Meeting Bots on Google Meet, Zoom, and Microsoft Teams.
Interested in the relationship technology/regulation/privacy, and open-source.
Our github: https://github.com/Meeting-Baas/meet-teams-bot
lazare [ _ at _ ] meetingbaas.com
Connected a few home cameras and two lights to an LLM, and made a few purchases.
The worst expensive offender being a tiny camera controlled RC Crawler[1]. The idea would for it to "patrol" my home in my name, with a sassy LLM.
1. https://sniclo.com/products/snt-niva-1-43-enano-off-road-803...
How long do contracts with such large corps usually take to enter in action, from the first meeting?
Yes we are a software service company selling that convenience. The code isn't "secret" but is definitely non-trivial to build and maintain.
My fear would rather be that a close-source competitor grabs our meeting bot, and builds on top, rather than using our service as an API.
If we were to open-source the software, we'd probably also offer it as an API for AI meeting companies.
We definitely need to improve the LP which has never been our forte.
As a team of developpers we spend way too much effort on the product itself, and not enough on marketing, the landing page, etc.
Which is ironic cause this little HN attention is thanks to our logo!
It’s the hand sign of “Spock” in Star Trek. The red point in the middle also stands for recording, and the green line + the red point stands for “happy person / user”
Having it automatically done is the next step but as we’re a generalist meeting tool it’s a tad harder (https://spoke.app)
We’ll done!
We provide the same functionnality (except for the Word export)
+ direct recording and upload from Zoom, Hangouts, etc.
+ video / audio editing & sharing by high-lighting which part of the transcript you'd like to keep.
www.spoke.app :)
In 70 languages (see language list here: https://spoke-for-sumo-lings.webflow.io/)
The idea behind it two-fold:
- people spend too much time in video-conferences as "by-standers" when receiving a summary of those conversations would be amply sufficient
- video-conferences transmit a ton of information that disappear into ether, or have to be written down.
First time I'm meeting my virtual Doppelganger, best part is, it's on HN !
Combined with signal-cli (https://github.com/AsamK/signal-cli.git) you can easily set up your own files & messaging upload. For instance, I've created an automatic upload to a Google Photos album for my family group chat and they love it !
On iOS Firefox and Chrome will actually also fail as they're all based on the Safari Web engine.
Sadly enough we can't make Spoke work with Safari yet as we're using video specific interpeted browser instructions.
Western leaders all seem to adopt the same compassionate tone, and wish to shine hope on the new year (with the exception of Trump).
On the other hand, Xi Jinping is just....self-congratulating.
Descript is an amazing product and one of the main inspirations for our startup and product : www.spoke.app
We offer the same services of editing by high-lighting text, but differ in so far as we offer video + direct capture of your content (both your microphone and system sound). Our goal isn't as much to produce nice, clean-cut videos, than to offer to summarize any video-conversation you might have for co-workers, friends, etc.
The goal was to summarize video conversations, save, and share them with our team easily. So we set out to build a tool to capture video-conversation moments, upload them, run Text-To-Speech solutions on them, and then provide a simple "video editing by text editing" interface to clean and share the videos in a few minutes.
We started by building the video recorder itself in Swift, with AVFondation. Swift has all the wanted features of a modern language, but XCode just ruins it. On top of that, the documentation of AVFondation has up to almost no documentation when you implement something out of the ordinary (for instance the function to choose the format of video recordings takes a dictionary where 99% of the values lead to a crash).
So to mitigate this issue, we now use ffmpeg to convert the movs created by AVFondation to a good audio and video format for the web and audio-analysis.
On the web side of things, we started out by using Python / Django. Frustrated by Django's monolith and poor documentation, we switched to Rust with Rocket, overcoming our fears of lack of maturity. Rocket has just the right level of abstraction, everything being explicit, with a lot of procedural macros to reduce the boilerplate. One limitation we however encountered was that Rocket does not handle well hanging connections, which can be very problematic when uploading videos. This ended up blocking all the threads if the user's connection was reset during an upload, so we had to put Nginx ahead of Rocket.
A few months after starting the journey of building Spoke we just launched version 1.0. If you're also bored by attending useless video-conference calls when somebody could've just sent you a summary, feel free to check it out.
[shameless plugin] I'm developing an app in a similar space (www.get-spoke.com) to be able to share video-conference summaries in seconds.
This would be a lot better than an NFC chip on a standard smartwatch as the F91W could last for years with that chip inside it.
The F91W is cheap, and really reliable. I've always fantasized about modding one, and this exactly would be my dream. Sadly, I don't have enough time on my hands.
If it were for sale, I'd buy it immediately !
For now, my team and I are focusing on video-conferences, but the end goal is much larger :)
Now that I think of it, this problem is a lot easier than what others post here.