Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Cracking Siri

applidium.com

Post reply on HN
Print view View original

91–100 of 105 posts

Re: Cracking Siri

#91

Post by david_a_r_kemp »

cstuder wrote: post #89
victoknight wrote: post #5

guess who doesn't verify the root CA. Think of all the fun to be had with a Siri man-in-the-middle

Have a look at the HTTP-proxy software Charles (No affiliation of mine.) In the last question of the FAQ the monitoring of SSL-connections within iPhone-apps is explained: http://www.charlesproxy.com/documentation/faqs/ There is no bug. This is what SSL will do, when you install additional certificates. (Oh, and it's a fun way to find new web services to play with.) :-)

Yeah - Fiddler2 http://fiddler2.com/fiddler2/ does this too. It's really neat for debugging http based API calls.

Re: Cracking Siri

#92

Post by rix0r »

kmm wrote: post #47
mbq wrote: post #36

Anyway, this is a proof that siri is a pure cloud service and as such may work even on 5-yo Sagem...

Not exactly. The text-to-speech is done in the cloud, but the hard part (algorithmically speaking) is natural language processing, which apparently (I don't know for sure) is still done on the phone. I don't know what Apple's excuse is though, but limited processing power is certainly not a problem.

I don't know why this is being downvoted -- I guess others are reading something different than I am?

There are three parts to Siri:

1. Speech-to-text (parent has it backwards but that's what he means, obviously)

2. Text-to-intent (referred to by parent as NLP)

3. Intent-to-API calls

Obviously, (1) happens in the cloud and (3) happens on the device. It is still unclear where (2) happens but if the cloud service only responds with text, it seems that (2) happens on the device.

And (2) is still a hard problem by itself.

Re: Cracking Siri

#93

Post by jrockway »

modeless wrote: post #90
jrockway wrote: post #88

Earlier quoted context omitted.

Why would that be problematic? You write a server that provides clients with an iPhone ID that hasn't been banned from using Siri yet, and then you make the app contact that server to get the ID. I'm sure Apple would send a nastygram, but they send nastygrams if you scratch your phone and don't get it repaired quickly enough. There is no law against telling other people your phone's serial number. There is no law aga…

Where would you get the valid IDs? You can't share the same ID between very many users, or Apple will ban it. You can't buy an iPhone for every user of your Siri app. iPhone users won't willingly give you their IDs. Are you going to somehow obtain and use the IDs of unsuspecting iPhone users without their permission? That is likely illegal and definitely will get you sued and booted from Android Market.

My guess is that Apple will ban an ID after a day or two. My other guess is that you can just keygen the ID.

Re: Cracking Siri

#94

Post by ugh »

alastairpat wrote: post #79
dfj225 wrote: post #76

Earlier quoted context omitted.

Sure, but there's nothing illegal about Apple kicking someone out of the App Store on a whim either. Doesn't matter if you are breaking the law or not, plenty of legal apps get rejected. Apple sets their own terms outside of US law.

Apple has to comply with its contract just as the developers do. I did just check the agreement and either party can terminate with 30 days' notice for any or no reason, so they could theoretically terminate. Given this is completely out of the scope of the App Store or even the SDK (contrast with the security researcher who got unapproved code executing), however, I don't imagine Apple will feel the need to terminat…

Apple usually avoids shitstorms or backpedals if they cause one – but sometimes they don't.

It's not unreasonable to assume at Apple won't do anything but it's risky.

Re: Cracking Siri

#95

Post by sovok »

philwelch wrote: post #85
mkjones wrote: post #75

Earlier quoted context omitted.

What makes you think thy've had a tough time scaling with early demand? I'd think that this scales horizontally pretty well, given that each request is largely stateless and there's no interaction between users.

> each request is largely stateless Negative. Siri remembers the context of your conversation.

But according to this article, the server seems to only translate speech to text, with the whole natural language processing and AI happening on the device.

Re: Cracking Siri

#96

Post by watty »

coyotej wrote: post #49
kmm wrote: post #47

Earlier quoted context omitted.

Not exactly. The text-to-speech is done in the cloud, but the hard part (algorithmically speaking) is natural language processing, which apparently (I don't know for sure) is still done on the phone. I don't know what Apple's excuse is though, but limited processing power is certainly not a problem.

I think you have it exactly backwards - the iPhone 4S/Siri speech-to-text/natural language processing are done in the CLOUD. The text-to-speech is done on the phone itself. My (non-Siri of course) iPhone 4's Voice Command stuff is COMPLETELY on the phone itself, and would do TtS of my contact list and Artist names, etc.

The article says, "The iPhone 4S really sends raw audio data". At least for Siri, TtS occurs on the cloud - not sure where the text processing > API occurs though.

Re: Cracking Siri

#98

Post by achompas »

I would LOVE to backward-engineer Siri's speech-analysis algorithms. Confidence scores help, but it doesn't look like any other modeling data is available?

Re: Cracking Siri

#99

Post by epo »

code_duck wrote: post #66
Volpe wrote: post #54

No one is at all concerned that this is a hack? I know it's interesting stuff, but I'm curious what "rights" Applidium have in publishing this information. With this information, (if I'm not wrong) it wouldn't take long to simply DDoS Siri... Or port Siri to Android (effectively stealing IP). (I have no bias either way, just pointing out, if someone figured out how to reverse engineer dropbox, so you could use their…

> 'No one is at all concerned that this is a hack?' You're asking that on a site called 'Hacker News' if I'm not mistaken. It is indeed a 'hack', a clever and skilled exploration of technology carried out with perfectly good or neutral intent.

The "hacker" part of the title is ironic. Any hacker news here is purely accidental.
Post reply on HN