or Connect
AppleInsider › Forums › Mobile › iPhone › Street test measures Siri comprehension at 83%, accuracy at 62%
New Posts  All Forums:Forum Nav:

Street test measures Siri comprehension at 83%, accuracy at 62% - Page 2

post #41 of 141
Quote:
Originally Posted by   View Post

If they add the expected larger 4"+ display along with LTE support, I completely agree. The next iPhone is probably a killer. 

Judging from the dominance of the current iPhone (leading seller, by all accounts) .... I'd have to say it's already a "killer".

See, in the record business, you can show someone your song, and they don’t copy it. In the tech business, you show somebody your idea, and they steal it. (Jimmy Iovine)
Reply
See, in the record business, you can show someone your song, and they don’t copy it. In the tech business, you show somebody your idea, and they steal it. (Jimmy Iovine)
Reply
post #42 of 141
Quote:
Originally Posted by DaHarder View Post


Then Apple should stop touting it as a primary 'feature' of the iPhone 4S in those ads/TV commercials until the 'feature' is complete i.e. Out of Beta.

DaHarder, I almost feel sorry for you. You keep trying to "badmouth" Apple .... but as each week passes and Apple keeps growing it's popularity, it's getting "DaHarder" and "DaHarder" for you to say anything at all .... that is, if you want to be truthful or relevant. Lucky for you that doesn't seem to be a priority. So sad.

See, in the record business, you can show someone your song, and they don’t copy it. In the tech business, you show somebody your idea, and they steal it. (Jimmy Iovine)
Reply
See, in the record business, you can show someone your song, and they don’t copy it. In the tech business, you show somebody your idea, and they steal it. (Jimmy Iovine)
Reply
post #43 of 141

siri is still in beta, so i'd say the figures are about right.

 

the one thing siri can do that google search can't is correct a majority of a speaker's pronunciation of their native language (when available in their country).

in this case, the pronunciation of the american english language has—not unlike it's citizenry—gotten fat, lazy and sloppy.

 

to me, that's priceless.

post #44 of 141
Quote:
Originally Posted by Rogifan View Post

How long is a service allows to be called beta? As long as its not working well we just call it beta to excuse away its problems.

We should revisit the "how long is a service allowed to be called beta" issue in another 4 1/2 years to see if Siri defeats Google's longest running beta record.
Edited by MacBook Pro - 6/29/12 at 9:07am
post #45 of 141
Quote:
Originally Posted by ankleskater View Post

Did I understand this correctly - Munster directly compared Siri to ... Google search by text input? 
 

 

Not only that, if you really wanted to see it's accuracy, you would have asked Siri to Google those queries, that way we can compared exactly how Siri compares to typed-input. The problem with this study is it is comparing two things at once. Siri's accuracy/comprehension, and Wolfram Alpha vs Google. How about making some of those queries specific to Wolfram's strengths and then we'll see just how good Google's searching is. Ask it some Math.

post #46 of 141
Quote:
Originally Posted by jragosta View Post


I guess critical thinking isn't your thing. I didn't object because of the result but because it was a stupid test. There were major flaws in his test:
1. Comparing a voice recognition system to manual text entry is inane.
2. His experiment is not reproducible because we don't know what queries he used and there is no control.
Once again, you obviously never took a critical thinking course. Nothing I said is grounds for any kind of lawsuit. What I said was:
"If someone uses a sample of native English speakers vs. a bunch of people just off the boat who learned English from a guidebook, the answer will be different."
That is a true, completely factual statement. I didn't say that anyone working for Munster fell into either category. I simply pointed out that voice recognition systems will depend on how similar the voice is to the average speaker for that language. If you want to do a real comparison, use the same speakers reading the same text to Siri and Google's voice recognition - and Siri comes out ahead
As to 'racist and xenophobic', that's one of the more bizarre comments you've ever made (and that's saying a lot). Simply stating that a voice recognition system will have more trouble understanding someone who doesn't speak good English is xenophobic? Pretty bizarre conclusion.

As obnoxious as your rant is, you do give rise to a valid question. If an English as a second language person was capable of communicating in English with a native speaker, then shouldn't one goal for Siri be the ability to also understand those same speech patterns? That to me would be a better test of the system's sophistication. We certainly have a wide range of regional accents even among native English speakers.

Life is too short to drink bad coffee.

Reply

Life is too short to drink bad coffee.

Reply
post #47 of 141
Quote:
Originally Posted by jragosta View Post

The problem is that this is a limited test and can not be easily verified. If someone uses a sample of native English speakers vs. a bunch of people just off the boat who learned English from a guidebook, the answer will be different.
The only real indication is what the millions of Siri users think about the product. One survey (using a random sample which should be representative) found that 96% if iPhone 4S users were somewhat satisfied or very satisfied with their phone and when asked about the best thing about the phone, 50% said 'Siri'.
http://www.uswitch.com/mobiles/news/2011/12/iphone_4s_siri_score_big_for_customer_satisfaction/
Or, you could do a side by side comparison using the same speakers and the same conditions. Siri had 96% accuracy compared to Google Voice at 93%:
http://blog.thearorareport.com/2011/11/14/aaple-siri-voice-recgnition-trumps-android-voice/
Of course, the real benefit is not just the accuracy, it's what you can do with it. If you're driving, there's some value in "Siri, make an appointment with John for 9 am next Wednesday". Even if you have to repeat it, it's better than having to rely on your memory until you get home or trying to manually enter the appointment while driving.

I seriously doubt Google will be able to provide a eyes-free, hands-free fully voice interactive experience for a long time. What many people don't realize is that Siri is the result of a decade of development by some of the greatest minds at some of the best technology schools in the United States at a total estimated cost of USD $200 million. A significant component in the development of Siri was a semantic web interface for scheduling events which is what enables Siri to communicate with various services such as Apple apps (Calendars, Contacts, Messages, Reminders) and especially external services such as movietickets.com, Wolfram Alpha, Yelp!.
post #48 of 141
Quote:
Originally Posted by robogobo View Post

It's "allowed" to be called beta as long as it's in beta.  And it's still in beta because they haven't worked out the problems completely.  As soon as they're worked out, it will probably be out of beta.  Make sense?

google mail was in beta for what seemed like forever!

post #49 of 141
Quote:
Originally Posted by jragosta View Post


ROTFLMAO.
Only in your world would a paid advertisement with no methodology or controls be a valid response to an independent side-by-side comparison with appropriate controls.
You know you've lost when you have to resort to an advertisement to make a claim that a product is better.

Dude, don't start out the day failing to read before posting, particularly when it's your own link. Only in your world could your article "proving" Siri more accurate than Google Voice Actions be misconstrued as a scientific study "with appropriate controls".  The article itself plainly says: 

 

"The study is small and not scientific, therefor no quantitative conclusions should be drawn

http://blog.thearorareport.com/2011/11/14/aaple-siri-voice-recgnition-trumps-android-voice/

 

That's your link!! Please be careful today. It's a long fall off your high-horse.

melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #50 of 141
Quote:
Originally Posted by mstone View Post

As obnoxious as your rant is, you do give rise to a valid question. If an English as a second language person was capable of communicating in English with a native speaker, then shouldn't one goal for Siri be the ability to also understand those same speech patterns? That to me would be a better test of the system's sophistication. We certainly have a wide range of regional accents even among native English speakers.

While that is certainly the goal, any thinking person realizes that a voice recognition system is going to fail when confronted with a speaker who is barely understandable.

A voice recognition system starts with the average voice and works from there. As it gets better, it learns to recognize voices that differ by more and more from the average, but under any conceivable system, a voice that is too far from average is going to be harder to recognize than an average voice.
Quote:
Originally Posted by Gatorguy View Post

Dude, don't start out the day failing to read before posting, particularly when it's your own link. Only in your world could your article "proving" Siri more accurate than Google Voice Actions be misconstrued as a scientific study "with appropriate controls".  The article itself plainly says: 

"The study is small and not scientific, therefor no quantitative conclusions should be drawn
http://blog.thearorareport.com/2011/11/14/aaple-siri-voice-recgnition-trumps-android-voice/

That's your link!! Please be careful today. It's a long fall off your high-horse.

And, yet, in the story I provided, they used multiple speakers and multiple different Android phones and read the same text to all of them (and the iPhones). So a comparison can easily be made. You may not like it and the author certainly is heading his bets, but it was a scientific survey.

Your citation was a Motorola advertisement that said that Motorola's phone was better.

Which do you think has more validity?
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
post #51 of 141
Quote:
Originally Posted by zeromeus View Post


Yes!  What a stupid test!  Of course the text input would be more accurate.  This isn't exactly news.


I wonder if he will compare it to sign language next?

post #52 of 141
Quote:
Originally Posted by Rogifan View Post


How long is a service allows to be called beta? As long as its not working well we just call it beta to excuse away its problems.
 

 

Quote:
Originally Posted by robogobo View Post

It's "allowed" to be called beta as long as it's in beta.  And it's still in beta because they haven't worked out the problems completely.  As soon as they're worked out, it will probably be out of beta.  Make sense?
 
 

 

Makes plenty of sense, but it doesn't address the issue. which is not how well Siri works but rather why are people NOT expecting to be beta testers when they get an iPhone.

 

It's all well and good that Apple still disclaims it on its Siri page on their site.   But I watched the SJ and ZD celebrity ads now just to check and there was no reference to it being in beta and no direction to go to the Apple site for clarification of its usage.  Apple is trying to play both sides of the fence, staying officially in beta while having a big public relations push with A-listers that doesn't mention it, as it has been in print ads I've seen.  

 

Obviously they're in the legal right and are not being outright deceptive, but some of those who are making objections to how it works aren't claiming it's not beta, but how this is/isn't presented.   To have the right to have a widely released, heavily advertised function be covered by the "beta" conditions it would seem you'd need to have it disclaimed on EVERYthing you use to promote it.   If that would be unwieldy or have a negative effect then you don't get to simply leave it off.  To me THAT's the issue, not that it's beta or that it's not perfect.

post #53 of 141
Quote:
Originally Posted by jragosta View Post

And, yet, in the story I provided, they used multiple speakers and multiple different Android phones and read the same text to all of them (and the iPhones). So a comparison can easily be made. You may not like it and the author certainly is heading his bets, but it was a scientific survey.
Your citation was a Motorola advertisement that said that Motorola's phone was better.
Which do you think has more validity?

The articles' authors (who are investment traders, possibly vested in Apple for all you know) say the study "was not scientific" and "no... conclusions should be drawn". Jragosta says the study was scientific and comparisons can easily be made. Who do you think has more validity, the guys that wrote the article or you?  I vote neither, tho the traders haven't yet proved to me that they're dishonest.  

 

Geesh, yet another instance where Jr can't be wrong. Yet here you are...


Edited by Gatorguy - 6/29/12 at 9:30am
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #54 of 141
Quote:
Originally Posted by lightknight View Post


So, basically since (for once) Munster has a result that doesn't proclaim Apple's dominance, his experiment is null and void? He explains how he did it, it's therefore verifiable. Moreover, your comment smells of racism and/or xenophobia. Let me highlight this for you: "bunch of people just off the boat who learned English from a guidebook". I _hope_ for you nobody on Munster's team is from foreign origin and willing to take you to court. Provided of course, they've got nothing better to do than read your comments, like me ;)

 

Munster has always been a tool.  He is the Digitimes of US Stock analysts.  Anything that comes from him can be immediately dismissed as churn fodder.  

 

Also take your threatening faux sense of superiority and hide it in your nether regions.  Your definition of racism/xenophobia is completely self serving and horribly misapplied.  While the post you responded to was a bit hyperbolic, it was not anything of which you describe.

.
Reply
.
Reply
post #55 of 141

I guess Siri will be in beta for years to come.

 

lsqgia26.jpg

post #56 of 141
Quote:
Originally Posted by jragosta View Post


1. Comparing a voice recognition system to manual text entry is inane.

Why is it inane? Other than you obviously not liking the result?

 

At the end of the day, they're both search engines. Siri translates the voice input to text and then submits it to a search engine. The accuracy of those searches was vastly in favor of Google. Deal with it.

 

Siri is a fun feature but its overall performance is lacking. It's great at the simple tasks, like setting alarms, providing directions home, voice dialing, but when you tell it to search for things it really does nose dive into uselessness.

post #57 of 141
Notably, SRI and Apple have only begun to demonstrate the currently available scope of Siri. Siri can do far more than anyone has seen in a released product today.

Siri can do Workflow Activity Recognition and Proactive Assistance (just like Google Now) today, this simply hasn't been activated by Apple.
Siri can handle email meeting requests, reserve venues, and schedule events then find emails, documents, and contacts that are relevant to a given meeting automatically, this simply hasn't been activated by Apple.
Siri can automatically compose email messages and suggest attachments for the message in response to scheduled events or emails, this simply hasn't been activated by Apple.
Siri can automatically search the Internet for contact information and additional data for your contacts, this simply hasn't been activated by Apple. For example, if you have a meeting with Susan Jeffries who is a high school friend you haven't seen for five years Siri can tell you she is married with two children (with their names and birthdays) and employed by Bank of America.
Siri can track your interests and generate a user-personalized RSS feeds, this simply hasn't been activated by Apple.

Siri can learn and automate complex workflows. To understand how powerful Siri is you need to understand how transformative the computer revolution was to businesses in the 1980s.

When Apple enables Siri on the Mac platform and enables use of these features, Siri will be very disruptive to business in a way that almost can't be understated. The problem is that this will be very disruptive to Apple as well since the flood of demand would be overwhelming. One might be forgiven for believing this might even be the beginning of the demise of the technology juggernaut, Microsoft.
Edited by MacBook Pro - 6/29/12 at 9:56am
post #58 of 141
Quote:
Originally Posted by jragosta View Post


While that is certainly the goal, any thinking person realizes that a voice recognition system is going to fail when confronted with a speaker who is barely understandable.
A voice recognition system starts with the average voice and works from there. As it gets better, it learns to recognize voices that differ by more and more from the average, but under any conceivable system, a voice that is too far from average is going to be harder to recognize than an average voice.

I've been using Google's voice recognition for quite a while and it works pretty well in both English and Spanish. I just started using Apple's and it seems to be on par with English but so far Spanish is not supported. I'm looking forward to the release later this year.

Life is too short to drink bad coffee.

Reply

Life is too short to drink bad coffee.

Reply
post #59 of 141

I think this is pretty impressive. My wife does not comprehend me and she is rarely right.  

post #60 of 141
Quote:
Originally Posted by MacBook Pro View Post

Notably, SRI and Apple have only begun to demonstrate the currently available scope of Siri. Siri can do far more than anyone has seen in a released product today.
Siri can do Workflow Activity Recognition and Proactive Assistance (just like Google Now) today, this simply hasn't been activated by Apple.
Siri can handle email meeting requests, reserves venues, and schedules events today then find emails, documents, and contacts that are relevant to a given meeting automatically, this simply hasn't been activated by Apple.
Siri can automatically compose email messages and suggest attachments for the message in response to scheduled events or emails, this simply hasn't been activated by Apple.
Siri can automatically search the Internet for contact information and additional data for your contacts, this simply hasn't been activated by Apple. For example, if you have a meeting with Susan Jeffries who is a high school friend you haven't seen for five years Siri can tell you she is married with two children (with their names and birthdays) and employed by Bank of America.
Siri can track your interests and generate a user-personalized RSS feeds, this simply hasn't been activated by Apple.
Siri can learn and automate complex workflows. To understand how powerful Siri is you need to understand how transformative the computer revolution was to businesses in the 1980s.

 

 

Got a link?  I know Siri was developed as a US  Government program, but I've never heard of these abilities.

"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #61 of 141
duplicate

Edited by Dick Applebaum - 6/29/12 at 1:16pm
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #62 of 141
Quote:
Originally Posted by jragosta View Post


Are you a flipping idiot or what?
I have a PhD in science. I know what is scientific.
Now, a scientific study may or may not have validity, but if it meets the premises of science, it is a scientific study. That is, it must have:
1. A testable hypothesis. Check
2. A methodology for testing the hypothesis. Check.
3. A control. Check.
4. Results. Check.
5. A conclusion. Check.
It meets the criteria. They are being careful to ensure that it's not being used as the be-all and end-all of research on the subject, but to say that it's not scientific just because it doesn't have a lot of replication is just plain wrong.
And, no matter how you slice it, it's infinitely more valid than a Motorola advertisement saying that Motorola's phones are better than Apple's phones. That should be obvious to anyone thinking at above a 5 year old's level. Sadly, that seems to leave you out.

Conclusion? The authors said there isn't a scientifically valid one. Controls?? What controls?? Were the readers aware of what phones were being used and thus intentionally/inadvertantly read to one device differently or at a higher/lower volume. Were all recordings at the same time or tested individually under varying conditions? Were they all new phones with properly working microphones? Was the noise level kept at a  constant level or more the traditional office rise and fall of voices, machinery and phones? Do you know the answers to any of that?

 

Sir, if you really have a PhD in science (which area BTW) you should use it as a basis for questioning rather than imagining the study you're so entralled with, likely done by a couple of guys with little/no scientific background in some day-trading office, was a "scientific study". You're doing your school a disservice.

 

Here's the "study" authors. They're stock traders for cryin' out loud, perhaps with vested interests in Apple for all you know. Scientists question. You sir are not a scientist, at least here, no matter what your piece of paper might or might not say.

http://thearorareport.com/


Edited by Gatorguy - 6/29/12 at 10:03am
melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #63 of 141
Quote:
Originally Posted by caliminius View Post

Why is it inane? Other than you obviously not liking the result?

 

At the end of the day, they're both search engines. Siri translates the voice input to text and then submits it to a search engine. The accuracy of those searches was vastly in favor of Google. Deal with it.

 

Siri is a fun feature but its overall performance is lacking. It's great at the simple tasks, like setting alarms, providing directions home, voice dialing, but when you tell it to search for things it really does nose dive into uselessness.


Such a comparison does not make sense. There's no way it would have been published in a peer-reviewed journal. A proper test would compare apples to apples, with control built in. If it makes sense to compare accuracy of speech processing to text input, why not include sign language and smoke signals too?

 

This has nothing to do with the utility of Siri. The reported test just does not pass any meaningful scientific scrutiny.

post #64 of 141
Quote:
Originally Posted by Gatorguy View Post

Controls?? What controls?? Were the readers aware of what phones were being used and thus intentionally/inadvertantly read to one device differently or at a higher/lower volume. Were all recordings at the same time or tested individually under varying conditions? Were they all new phones with properly working microphones? Was the noise level kept at a  constant level or more the traditional office rise and fall of voices, machinery and phones? Do you know the answers to any of that?

 

Sir, if you have a PhD in science (which area BTW) you should use it as a basis for questioning rather than imagining the study you so entralled with, likely done by a couple of guys with little/no scientific background in some day-trading offic,e was a "scientific study". You're doing your school a disservice.

 

Here's the "study" authors. They're stock traders!

http://thearorareport.com/


Indeed, their profession makes them ill-qualified to conduct proper testing. But, in this case, the comparison does not even pass the test of common sense.

post #65 of 141
Quote:
Originally Posted by jragosta View Post


Are you a flipping idiot or what?
I have a PhD in science. I know what is scientific.
Now, a scientific study may or may not have validity, but if it meets the premises of science, it is a scientific study. That is, it must have:
1. A testable hypothesis. Check
2. A methodology for testing the hypothesis. Check.
3. A control. Check.
4. Results. Check.
5. A conclusion. Check.
It meets the criteria. They are being careful to ensure that it's not being used as the be-all and end-all of research on the subject, but to say that it's not scientific just because it doesn't have a lot of replication is just plain wrong.
And, no matter how you slice it, it's infinitely more valid than a Motorola advertisement saying that Motorola's phones are better than Apple's phones. That should be obvious to anyone thinking at above a 5 year old's level. Sadly, that seems to leave you out.

 

What was the hypothesis? That one phone will perform better? What predictions were involved, and what new phenomena is now capable of being explained by the hypothesis that was not capable of being explained previously, hence requiring this hypothesis?

post #66 of 141
Quote:
Originally Posted by kent909 View Post

I think this is pretty impressive. My wife does not comprehend me and she is rarely right.  


But, unlike your wife, you would dare to tell Siri that she was wrong. :)

post #67 of 141
Quote:
Originally Posted by Gatorguy View Post

The articles' authors (who are investment traders, possibly vested in Apple for all you know) say the study "was not scientific" and "no... conclusions should be drawn".

 

These are perfunctory (even legal?) liability-shedding statements. The fact is that they publish these reports aimed to planting conclusions in people's heads. Otherwise, what purpose does the report serve? 

post #68 of 141
Quote:
Originally Posted by ankleskater View Post

 

These are perfunctory (even legal?) liability-shedding statements. The fact is that they publish these reports aimed to planting conclusions in people's heads. Otherwise, what purpose does the report serve? 

I completely agree. They had some purpose in mind by publishing it, and it wasn't in the interest of science IMHO.

melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
post #69 of 141

SIRI and Apple's dictation is not 100% perfect of course, but it works pretty good, and it's obviously only going to get better as Apple keeps on improving it.

 

It seems to me that most of the people who have problems with SIRI are people with speech defects, foreigners with ridiculous accents and illegal aliens. Have you ever seen some of the youtube videos of certain people trying to use SIRI? SIRI is not the problem, it's those people and their laughable voices along with their incomprehensible and embarrassing pronunciations.

post #70 of 141
Quote:
Originally Posted by mstone View Post

I've been using Google's voice recognition for quite a while and it works pretty well in both English and Spanish. I just started using Apple's and it seems to be on par with English but so far Spanish is not supported. I'm looking forward to the release later this year.

That's an entirely different issue.

It should recognize properly spoken English better than poorly spoken English. Similarly, it would recognize properly spoken Spanish better than poorly spoken Spanish. It's ability to recognize multiple languages doesn't mean that it will be capable of recognizing words that are garbled and mis-spoken in any given language.
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
"I'm way over my head when it comes to technical issues like this"
Gatorguy 5/31/13
Reply
post #71 of 141
Quote:
Originally Posted by Gatorguy View Post

The articles' authors (who are investment traders, possibly vested in Apple for all you know) say the study "was not scientific" and "no... conclusions should be drawn". Jragosta says the study was scientific and comparisons can easily be made. Who do you think has more validity, the guys that wrote the article or you?  I vote neither, tho the traders haven't yet proved to me that they're dishonest.  

 

Geesh, yet another instance where Jr can't be wrong. Yet here you are...

 

Although, regardless of how "scientific" this study was,  it certainly carries more validity than GG's Motorola ad. Seriously, GG, that's an all new low even for you to try to use an ad as proof of such a claim.

 

However, although I haven't looked at jragosta's cited study, disclaimers by the authors that it wasn't scientific, or that no conclusions should be drawn don't necessarily invalidate it. It's possible that their study was methodologically correct with significant results, but either a) they didn't realize it was or b) as investment traders they made disclaimers as a matter of course so no one could sue them because investment decisions based on the results didn't turn out well. Even professional scientists will sometime caution against basing action on the results of seemingly valid studies.

post #72 of 141
Quote:
Originally Posted by "Apple 
[" url="/t/151000/street-test-measures-siri-comprehension-at-83-accuracy-at-62/40#post_2137151"]SIRI and Apple's dictation is not 100% perfect of course, but it works pretty good, and it's obviously only going to get better as Apple keeps on improving it.

It seems to me that most of the people who have problems with SIRI are people with speech defects, foreigners with ridiculous accents and illegal aliens. Have you ever seen some of the youtube videos of certain people trying to use SIRI? SIRI is not the problem, it's those people and their laughable voices along with their incomprehensible and embarrassing pronunciations.

My wife is born and raised Japanese, listening to her trying to coax information from Siri in English is hysterically funny. Siri in Japanese seems to be fairly reliable for her though.
post #73 of 141
Quote:
Originally Posted by MacBook Pro View Post


My wife is born and raised Japanese, listening to her trying to coax information from Siri in English is hysterically funny. Siri in Japanese seems to be fairly reliable for her though.

We have something in common, because my gf is also Japanese. :)

post #74 of 141
Quote:
Originally Posted by Gatorguy View Post

I completely agree. They had some purpose in mind by publishing it, and it wasn't in the interest of science IMHO.


Definitely not for "scientific" purposes but rather for promotional purposes. After that, these folks promote themselves as "researcher analysts". With that job title, they do need to provide rigorous analysis, just as pollsters do.

post #75 of 141

you are one person

Citing unnamed sources with limited but direct knowledge of the rumoured device - Comedy Insider (Feb 2014)
Reply
Citing unnamed sources with limited but direct knowledge of the rumoured device - Comedy Insider (Feb 2014)
Reply
post #76 of 141
Quote:
Originally Posted by Apple ][ View Post

It seems to me that most of the people who have problems with SIRI are people with speech defects, foreigners with ridiculous accents and illegal aliens. Have you ever seen some of the youtube videos of certain people trying to use SIRI? SIRI is not the problem, it's those people and their laughable voices along with their incomprehensible and embarrassing pronunciations.

 

Quote:
Originally Posted by Apple ][ View Post

We have something in common, because my gf is also Japanese. :)

Does she know how you feel about foreigners?

Life is too short to drink bad coffee.

Reply

Life is too short to drink bad coffee.

Reply
post #77 of 141
Quote:
Originally Posted by jragosta View Post


That's an entirely different issue.
It should recognize properly spoken English better than poorly spoken English. Similarly, it would recognize properly spoken Spanish better than poorly spoken Spanish. It's ability to recognize multiple languages doesn't mean that it will be capable of recognizing words that are garbled and mis-spoken in any given language.

What issue did you read into my post? There is no issue.

Life is too short to drink bad coffee.

Reply

Life is too short to drink bad coffee.

Reply
post #78 of 141
Quote:
Originally Posted by mstone View Post

 

Does she know how you feel about foreigners?

 

She most certainly does. And I don't have anything in particular against foreigners, in most cases. I'm just pointing out that people who speak ridiculously have no right to complain about the accuracy of a computerized speech recognition program. If I tried to give commands in Spanish or Japanese or German, that would most definitely sound pretty ridiculous too.

post #79 of 141

When I was in highschool many moons ago, understanding the teacher 86% of the time and only getting 62% of my answer's right was a big fat F, of course I just wrote BETA on the top of my tests and all was forgiven.

post #80 of 141
Quote:
Originally Posted by mstone View Post

 

Does she know how you feel about foreigners?

 

He doesn't pronounce it fur'ners??  Just kidding... lol.gif

melior diabolus quem scies
Reply
melior diabolus quem scies
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: iPhone
AppleInsider › Forums › Mobile › iPhone › Street test measures Siri comprehension at 83%, accuracy at 62%