4 ms·

One startup strives to make Iron Man's J.A.R.V.I.S. a reality

10 points by prateekj 13y ago

cromwellian 13y ago
Hate to be negative, but not really all that impressed. JARVIS implies AI and conversational/contextual UI, this just appears to be a thin voice action skin which has been done for decades, not sure it was worthy of a Forbes article.
- kybernetikos 13y ago
  I know that JARVIS in Iron Man was an AI, but actually when I watched Iron Man, I was particularly impressed with just how well integrated the voice system was and how if a startup put their mind to it, they probably could achieve something close with todays technology: A house wide audio system that recorded all speech, fingerprinting it so that the speaker was identified and transcribing it so that it was searchable. Add to that a SIRI-like system that you can interact with naturally from any room in your house, and you already have something very interesting. Just think about how useful it would be for solving domestic disputes: "Jarvis, did I agree to buy the milk yesterday?", "Could not find any conversations involving you and 'milk' yesterday." :-)
- ddeck 13y ago
  >not sure it was worthy of a Forbes article. Don't confuse Forbes magazine with Forbes.com. The website uses the same model as sites like seeking alpha, where pretty much anyone can write poorly researched, unverified articles and get them published. From Wikipedia: In contrast to the Forbes magazine, the website Forbes.com uses a "contributor model" in which a wide network of "Contributors" writes and publishes articles directly on the website.[15] The Forbes staff does not assign stories, fact check, or edit contributions, and Contributors write stories about any topic they choose.[15] Contributors are paid based on traffic to their Forbes.com pages; the site has received contributions from over 2,500 individuals, and some Contributors have earned over US$100,000, according to the company.[15]
nsxwolf 13y ago
I started watching the video. By 2:30, I had seen a voice recognition Google search, and what appeared to be a voice recognition "I'm Feeling Lucky" Google search. Bored with this, I started skipping around looking for the really whiz-bang AI. I saw some home automation stuff that didn't seem to be much more than something OS X Automator could do by kicking off a script. If you roll enough dumb voice activated services together, that can still be a useful tool, but I didn't see anything that was at Siri's level. Maybe it's hidden in the video somewhere. If it is, it should have been showcased right away.
- austinl 13y ago
  Same exact experience - I was at least hoping for some cool natural language processing. Their "generate reports" feature just scrapes text from Wikipedia...
hipaulshi 13y ago
The Startup I am currently working on (also called "Jarvis Inc"), is aiming for the same star. While the current form is far from the contextual understanding. I love to see people working towards the same goal. This was our earlier mockup at jarvis.co. We only saw limited interest probably because of the high end market size. We are currently in the middle of building an open API platform as a spin off of this project.
electic 13y ago
Hate to be an a-hole but this is not impressive at all. The UI, the interface, and UE is woefully deficient compared to SIRI or offerings from Google.
- jsumrall 13y ago
  I don't mind. Their OS sucks. But I think it was the forbes author who pumped up our expectations.
thenerdfiles 13y ago
You have to click an icon to activate voice... I'm sure you can write a quick Python script to compare new audio input, stored in a watched directory, to a base audio dictionary that fires an API based on normalized matches to "terms" in that dictionary. If you say "search" the script compares/approximates to existing waveforms. We should have a Speech-Computer Interface API, rather than boring ourselves with the standard "Search for...", that makes a canonical set of speech patterns which computers should respond to habitually. Things like "I'm thinking..." (fires off search, feeling lucky, wikipedia, etc. mashup results pages), "I'm hungry..." (fires off search, geolocation information, personal biometric data, etc...), "Who is..." (wiki? pipl? facebook?), "What is..." (wiki?), "When is..." (shows a calendar)...
obilgic 13y ago
I am not feeling lucky for the people behind this...

10 points by prateekj 13y ago

Hate to be negative, but not really all that impressed. JARVIS implies AI and conversational/contextual UI, this just appears to be a thin voice action skin which has been done for decades, not sure it was worthy of a Forbes article.

kybernetikos 13y ago
I know that JARVIS in Iron Man was an AI, but actually when I watched Iron Man, I was particularly impressed with just how well integrated the voice system was and how if a startup put their mind to it, they probably could achieve something close with todays technology: A house wide audio system that recorded all speech, fingerprinting it so that the speaker was identified and transcribing it so that it was searchable. Add to that a SIRI-like system that you can interact with naturally from any room in your house, and you already have something very interesting. Just think about how useful it would be for solving domestic disputes: "Jarvis, did I agree to buy the milk yesterday?", "Could not find any conversations involving you and 'milk' yesterday." :-)
ddeck 13y ago
>not sure it was worthy of a Forbes article. Don't confuse Forbes magazine with Forbes.com. The website uses the same model as sites like seeking alpha, where pretty much anyone can write poorly researched, unverified articles and get them published. From Wikipedia: In contrast to the Forbes magazine, the website Forbes.com uses a "contributor model" in which a wide network of "Contributors" writes and publishes articles directly on the website.[15] The Forbes staff does not assign stories, fact check, or edit contributions, and Contributors write stories about any topic they choose.[15] Contributors are paid based on traffic to their Forbes.com pages; the site has received contributions from over 2,500 individuals, and some Contributors have earned over US$100,000, according to the company.[15]

nsxwolf 13y ago

I started watching the video. By 2:30, I had seen a voice recognition Google search, and what appeared to be a voice recognition "I'm Feeling Lucky" Google search. Bored with this, I started skipping around looking for the really whiz-bang AI. I saw some home automation stuff that didn't seem to be much more than something OS X Automator could do by kicking off a script. If you roll enough dumb voice activated services together, that can still be a useful tool, but I didn't see anything that was at Siri's level. Maybe it's hidden in the video somewhere. If it is, it should have been showcased right away.

austinl 13y ago
Same exact experience - I was at least hoping for some cool natural language processing. Their "generate reports" feature just scrapes text from Wikipedia...

hipaulshi 13y ago

The Startup I am currently working on (also called "Jarvis Inc"), is aiming for the same star. While the current form is far from the contextual understanding. I love to see people working towards the same goal. This was our earlier mockup at jarvis.co. We only saw limited interest probably because of the high end market size. We are currently in the middle of building an open API platform as a spin off of this project.

electic 13y ago

Hate to be an a-hole but this is not impressive at all. The UI, the interface, and UE is woefully deficient compared to SIRI or offerings from Google.

jsumrall 13y ago
I don't mind. Their OS sucks. But I think it was the forbes author who pumped up our expectations.

thenerdfiles 13y ago

You have to click an icon to activate voice... I'm sure you can write a quick Python script to compare new audio input, stored in a watched directory, to a base audio dictionary that fires an API based on normalized matches to "terms" in that dictionary. If you say "search" the script compares/approximates to existing waveforms. We should have a Speech-Computer Interface API, rather than boring ourselves with the standard "Search for...", that makes a canonical set of speech patterns which computers should respond to habitually. Things like "I'm thinking..." (fires off search, feeling lucky, wikipedia, etc. mashup results pages), "I'm hungry..." (fires off search, geolocation information, personal biometric data, etc...), "Who is..." (wiki? pipl? facebook?), "What is..." (wiki?), "When is..." (shows a calendar)...

obilgic 13y ago

I am not feeling lucky for the people behind this...