Slashdot Mirror
Microsoft's Acoustic Caller ID Patent
theodp writes "A new patent granted to Microsoft Tuesday for automatic identification of telephone callers based on voice characteristics
covers constructing acoustic models for telephone callers by identifying words or subject matter commonly used by callers and capturing the acoustic properties of any utterance. Not only that, it's done 'without alerting the caller during the call that the caller is being identified,' boasts Microsoft in the patent claims."
The only difference here (aside from what agencies have been doing since the 1960's) is that this analysis seems to be done in real time, rather than offline? I mean, haven't monitoring people been able to tell who is speaking based on sound synthesis since forever?
Anecdotally I feel like some companies answer the phone quicker if you talk to their automated system in an irate and condescending manner. Could just be me though :)
What's the purpose of caller ID after I've picked up the phone? I'm not going to talk to some challenge response bot if I'm someone who needs to be IDd and screened anyway.
We've upped our standards. Up yours.
Brilliant!
Quo usque tandem abutere, Nimbus, patientia nostra?
I read the patent, but I guess I don't get it. How is what Microsoft claiming to do different from existing voice recognition systems?
You have to train current voice systems so they recognize your voice pattern (or, acoustic ID) and translate it to text or action. Take that and add a system that keeps profiles for a more advanced version of caller ID. It seems like a natural evolution of the technology.
Bearded Dragon
If someone had acquired some of your personal information, and then tried to impersonate you, an automated voice recognition system could be useful by raising an alarm, or at least giving a percentage of how much their voice is like yours.
Wont this most likely violate wiretapping laws in two-party states?
Your hair look like poop, Bob! - Wanker.
I had no idea someone I might call might be able to indentify me.
The sort of processing this patent covers is something that hasn't been possible until recently, but I think, in principle, is something absolutely necessary for robust AI, and that is doing recognition simultaneously on both low level features and high level features of data and on intersections of the two.
By "high level" I mean things like word choice, language etc. By low level I imagine they mean things like the specific resonance characteristics of a voice. In voice there are intermediate levels of features too, such a the characteristics of phonemes.
The upshot of this is that just as algorithms and hardware begins to reach a level of power necessary to show intelligence, it will be impossible to do so without stepping on patents.
We will have patents on a machine not being stupid.
Comment removed based on user account deletion
/. should just put an RSS feed to newly issued patents on the front page. Would cut down on the number of stories per day though.
It is my understanding that recording a telephone conversation is against the law in most states, without notifying the other parties on the line.
Thus, a practical device for this patent would most likely be illegal.
... but it works as well as their speech recognition. Between this, face recognition and kill drones OBL will be found and exterminated early and often. I hope it's not me next.
The programmers put in an Easter egg, just for you. Whenever Twitter says "shit" into his cell phone, the official Microsoft transcript has "M$".
Friends don't help friends install M$ junk.
To patent anything, follow these steps:
1. Choose something already being done in the real world, anything really
2. describe it with maximum verbosity
3. add "on the Internet" at the end
Tada! PATENT!
Inventors: Arthur C. Clarke and Stanley Kubrick
First publication: 2001 A Space Odyssey (Released 1968). Heywood Floyd checks in to the space station:
Female voice: "Thank you. You are cleared through Voiceprint Identification."
http://www.imdb.com/title/tt0062622/quotes
NSA has had real-time voice ID since before '96 and possibly longer. How MS got this patent is beyond me. Our system is soooooooo broken
I prefer the "u" in honour as it seems to be missing these days.
According to this:
;)
Not only that, it's done 'without alerting the caller during the call that the caller is being identified,'
They are describing a means to RECORD callers without their knowledge, and hence without their consent. So would this software be illegal in some jurisdictions? You bet yer ass it would be.
Wonder how it handles people who say "uhm" or "uhh" a lot.
My Suburban burns less gasoline than your Prius.
My name is Werner Brandes, my voice is my passport. Verify me.
"How is what Microsoft claiming to do different from existing voice recognition systems?"
Existing voice recognition systems might be more acurately called speech recognition. They don't recognize the voice (who is speaking); they recognize the speech (what is being said). They can be categorized as speaker dependent or speaker independent.
Speaker dependent speech recognition (type 1) requires complex training by each user. It needs to know all the ways a person pronounces every possible phoneme. During use, it must be given the name of the speaker and a sound sample. It gives back the name of the phoneme. 2 inputs, 1 output.
Speaker independent speech recognition (type 2) is able to identify individual phonemes as spoken by a wide variety of speakers. 1 input, 1 output. That's what I would imagine is the important first step of what MS is claiming to do. Once a phoneme or two has been identified, the name of the phoneme and the captured sound sample can be fed to the type 1 algorithm and it would be able to output the name of the speaker.
Functionally it's different than existing "voice recognition" systems, but I seriously doubt it worthy of a patent.
...it's done 'without alerting the caller during the call that the caller is being identified.
...Sometimes...when the phone rings...
...I answer it...and just listen...
...I hear the caller's voice and identify them by their voice...
...Then hang up without saying anything.
How insidious!
What. Is. The. Difference.
"The hot chick with the Uzi... can I get her phone number?"
"So after all this, you make my case for me. To end this stalemate, you must die..."
I have Caller ID so I know who's calling BEFORE I pick up the phone, not afterwards.
"We need to get over this notion, that, for Apple to win... Microsoft must lose." - Steve Jobs, 1997
Should I even ask? Does the 4th Amendment mean anything anymore?
Cops bust a guy for video taping them and charge him with wiretapping and Microsoft is going to be recording my voice and compiling a profile of me and that's okay?
Words I'm guessing it will be looking for by default: bomb, liberal, weed, nuke, bush, 1st Amendment.
My tinfoil hat is starting to look stylish.
me: hello?
caller: Hello, I'm Suzi Cheatem from Dewey, Cheatem, and Howe. I was wondering if you'd like to answer a few questions about your behaviour while using the Internet.
I think hrm, this sounds like one of those annoying telemarketers
me: Sorry, I'm not interested in speaking to telemarketers
caller: It seems like you have identified me from a previously identified acoustic model. I'm afraid I'm going to have to tell Microsoft that you have stolen their idea. You can expect a bill from them within two weeks.
<hangs up>
Gosh, those telemarketers get stranger every time they call me.
Ask me about repetitive DNA