Posts tagged with "siri"

The Siri API→

The Siri API

Samuel Iglesias has written an excellent post detailing the (possible) challenges developers will have to cope with if Apple decides to release a Siri API.

The second half of Siri integration, Semantics, is the tricky part: something that most iOS developers have never dealt with. Semantics will attempt to capture the various ways a user can ask for something, and, more importantly, the ways Siri, in turn, can ask for more information should that be required. This means that developers will need to imagine and provide “hints” about the numerous ways a user can ask for something. Sure, machine learning can cover some of that, but at this early stage Siri will need human supervision to work seamlessly.

This is exactly what I have been wondering since speculation on the Siri API started last year. How will an app be capable of telling Siri the kinds of input (read: natural language) it accepts? Will developers have to do it manually? Will Apple provide a series of automated tools to associate specific features (say, creating a task in OmniFocus) with common expressions and words? And how is Apple going to look into the natural language processing developers will implement in their apps?

Of course, the Siri API is still at the speculation stage, but it does make sense to greatly expand upon Siri’s capabilities as an assistant capable of working with any app. The TBA sessions at WWDC are intriguing, and Tim Cook said we’ll be pleased with the direction they’re taking with Siri. Right now, I’d say integrating with third-party software would be a fantastic direction.

Permalink

Apple Posts New Siri Ads Featuring John Malkovich

Following a series of “celebrity ads” for the iPhone 4S’ voice-based assistant released last month, Apple today posted two new Siri TV commercials featuring actor John Malkovich. The ads, titled “Joke” and “Life” show Malkovich casually talking to Siri with short sentences and a series of single words such as “weather” or “evening”, perhaps in an effort to showcase both Malkovich’s particular attitude and Siri’s capability of handling short commands with seemingly no context (“evening” returns a series of calendar appointments, “linguica” displays local restaurants).

According to a recent study, the previous commercials featuring Zooey Deschanel and Samuel L. Jackson fared well with viewers, who, reportedly, were highly receptive to familiar faces of celebrities illustrating the latest features of the iPhone in a familiar, almost casual setting. MacRumors has put together a number of possible responses Siri can give to Malkovich’s query – tests performed with the question asked by Samuel L. Jackson showed that, in practice, Siri was a little less accurate than its primetime counterpart.

The new Siri ads are available on Apple’s website, YouTube channel, and we have embedded the official versions below.
Read more

iTranslate Voice: A Siri-like Translator Powered by Nuance

According to Apple, Siri, the iPhone 4S’ virtual voice-based assistant, will gain support for additional languages in 2012. While international users are still waiting for Siri to support the creation of reminders, events, and messages in their preferred language – not to mention all the other functionalities that Apple enabled in the first “beta” of the assistant – it is still unclear if Siri will ever officially support a feature that has always seemed perfect for voice input and interactions: translations. iTranslate Voice, a new utility by Sonico Mobile, wants to fill this void with a Siri-like interface for an iPhone app powered by the same tech behind Apple’s solution: Nuance.

If Siri could do voice translations, I imagine they would look exactly like iTranslate Voice, as Sonico’s app borrows heavily from Siri in terms of overall interface design and style. Like Siri, iTranslate Voice displays ongoing voice interactions as conversations between you and the software; like Apple’s assistant, conversations are initiated by tapping on a circular, glowing microphone button that, however, in this app has been styled with flags. Conversations can be scrolled vertically, and tapping on single “bubbles” will you give you options to copy, share, or “speak” the words you dictated again. By tapping and holding the latest entry in a conversation, you can directly modify the words you said using the iPhone’s keyboard – this can be particularly useful to refine text input in case the app didn’t get some details right.

iTranslate uses Nuance’s speech recognition software to recognize and translate voices. iTranslate supports 31 languages, but only some of them support both voice recognition and text-to-speech. A complete list of supported languages (which also includes dictionary definitions, automatically displayed inline when available) is available on the developers’ website. Like Siri, iTranslate Voice requires an Internet connection to operate – something that, I presume, is related to some heavy server-side processing and sampling the app does in order to deliver high-quality and timely results without wasting the iPhone’s local storage.

In my tests, iTranslate Voice has been extremely accurate and reliable. I was surprised at first, but on second thought I realized that the fact that Nuance is delivering good translations in seconds as you talk to your phone shouldn’t really be a surprise at this point. I have tested the app with Italian-to-English translations (and vice versa), and the results have been more than decent – I actually think they are the best ones I have found on iOS to date, and this speaks clearly of Nuance’s strong position in the market. I don’t know if the developers are also enhancing Nuance’s results with their own engine of sorts, but the end result is what matters, and iTranslate Voice doesn’t disappoint here. The app recognized first names and terms like “ATM”, handled common expressions well, and it even understood spoken punctuation in Italian (“virgola” for comma, “punto interrogativo” for question mark, and so forth), adjusting the sentences and tone of software accordingly. Results are delivered in seconds both on WiFi and 3G, and the app also does a good job at detecting end of speech if you enable the option in the Settings.

I won’t judge Sonico’s decision to make iTranslate Voice look like Siri, but I will say that the system undoubtedly works, and makes it easy to speak to your phone to get instant, spoken translations. I can’t shake the feeling that this, like Siri, feels like the future of human-to-computer interactions being built right now, and regardless of whether Apple and Nuance will eventually bring this feature to Siri, iTranslate Voice is impressive and you can get it today.

iTranslate Voice is available on the App Store at $0.99.

Samuel L. Jackson and Siri Star in New iPhone 4S Commercial (Update: Zooey Deschanel Too)

As noted by The Next Web, U.S. carrier Verizon Wireless aired a new iPhone 4S commercial featuring actor Samuel L. Jackson and Siri, the company’s virtual assistant for the iPhone 4S. Remarkably similar to Apple’s own ads in terms of style and message, the ad places Siri front and center, showcasing various functionalities of the software. Samuel L. Jackson asks his assistant to cancel appointments, create reminders, look up locations, and convert units to prepare for his “date night”. At the end of the commercial, the actor ironically asks Siri to take the night off, to which Siri replies “if you say so”.

Even more ironically, as depicted in the screenshots above, Siri really does reply to that command with a series of different answers.

Verizon Wireless’ official YouTube channel hasn’t been updated yet with the new commercial, and it’s not clear whether Apple’s official ad agency Chiat\Day may have been involved in the creation of the ad. You can check out an official embed after the break.

Update: It appears a full “Siri and celebrities” ad campaign is going live today. Another ad featuring actress Zooey Deschanel just aired as well, for carrier Sprint. It also seems like the same ads are airing with different carriers logo at the end, confirming the Samuel L. Jackson ad wasn’t simply a Verizon Wireless commercial, as initially suggested.

Update 17/4: Apple just posted the official versions of the ads on its website and YouTube channel. Find them below, or on Apple’s website (Date Night, Rainy Day).
Read more

“Okay, I’ll Remind You”

A few minutes ago Apple uploaded two new iPhone 4S commercials on its website and official YouTube channel. The ads, as with previous iPhone 4S promotional videos, focus on Siri, and they might just be the best ones about the voice-based assistant yet.

The ads, called “Rock God” and “Road Trip”, share a common theme: people talking to their assistant using natural language and a friendly tone, not simply asking a piece of software to execute commands.

In Road Trip, a guy and his girlfriend are organizing a road trip to California. Look at the initial setting: it’s cold outside, they’re about the get in the car, and they want to get from the cold of East Coast in February to the sunny Santa Cruz in California. The guy asks Siri, and they’re on the road. Camera cuts to the guy’s face in the car. He’s looking for a barbecue in Kansas City. Camera cuts to girl’s face in the car. She’s looking for a rodeo. Camera changes again, this time the couple doesn’t know where they are, and the girl asks “Where are we?”, with the look of someone who knows Siri will have an answer. They’re in Santa Rosa. Change again. How big is the Grand Canyon? Sure enough, Siri can look that up on Wolfram Alpha or Wikipedia. But then the gas runs out: how about finding a station the guys can walk to? Finally, the ad reaches its climax when our two characters have seemingly reached their destination, or are fairly close, and are looking at the stars. She asks: “What does Orion look like?”. Siri displays sky data inline. The video closes with the opposite setting of how it began: sunny California, he’s wearing a t-shirt, looking at the horizon, and she’s telling Siri like you would do with an old friend – Remind me to do this again. Siri, with its human-like voice, replies: Okay, I’ll remind you.

The second commercial, Rock God, has a more “fun” approach. There’s this kid that “has to get a guitar”, and he’s so excited about it he needs to ask Siri now. Why is he so excited about getting a guitar? What’s the story here? Perhaps, I imagine, he has just decided with his friends that it’s time to put a band together and start playing. So, Siri gives him location info about stores selling musical instruments, and in the next scene our kid is learning how to play. How do I play London Calling? Whole Lotta Love? How about that chord? Siri displays information on screen. Fairly regular stuff for now. Then the ad changes – our character is sending a message to Julie and Kate about playing at the garage tonight. Apple’s music stops. The kids are playing – they’re doing rock ‘n’ roll! – and finish their song. “Call me Rock God”, the kid tells Siri, softly.

You see, these aren’t just ads. In 30 seconds, we’re told stories. In 30 seconds, we are not shown technical features and RAM specifications, we’re given real examples of real people we can relate to. We’re shown two young people in love with each other that just want to get to California and see the sunset together. We’re shown a young boy with a simple dream, playing guitar, yet a dream that’s important to him – something that makes his life worth living and enjoying even for those 30 minutes when he gets his band mates together and nothing else matters. Just music. Call me Rock God.

In 30 seconds, we’re shown how technology can make people’s lives better. We’re reminded, once again, that this industry, this love for the latest gadget, doesn’t necessarily have to be about tech specs – it’s the technology married with the liberal arts. It’s about playing London Calling with your friends. It’s about driving to California with the woman you love and watch the stars just for one night.

He would be proud.

Read more

Siri Accounts for 25% Of Wolfram Alpha Queries→

Siri Accounts for 25% Of Wolfram Alpha Queries

Following last night’s announcement of Wolfram Alpha Pro, coming tomorrow at $4.99 per month, The New York Times’ Steve Lohr reported in his piece about Siri’s integration with the service. Apple’s voice-based assistant accounts for 25% of all Wolfram Alpha searches four months after the launch of the iPhone 4S, which embeds Wolfram-powered results directly into the Siri interface.

The subjects in the Wolfram Alpha database are now more useful to the average person. Type in “Tinker Tailor Soldier Spy showtimes,” and Wolfram Alpha delivers the schedule for local theaters. The movie times, Dr. Wolfram notes, come not from scouring the Web, but from a specialized information service.

Siri accounts for about a quarter of the queries fielded by Wolfram Alpha, whose staff has grown to 200. Several large companies in health care, financial services and oil and gas recently hired Dr. Wolfram’s private company, Wolfram Research, to do tailored corporate versions of Wolfram Alpha for them. Microsoft also licenses Wolfram Alpha technology.

With 37 million iPhones sold in the last quarter and the iPhone 4S being the most popular model among them, I assume a good percentage of new 4S owners are using Siri regularly to file requests through Wolfram’s knowledge engine. The question is, how regularly? Whilst some have already dismissed Siri as a “passing fad”, I’m hearing of people using it on a daily basis for things like quick reminders, alarms, and data lookups (queries like “how many people live in France?” are perfect for Wolfram Alpha). Saying Siri is not a fad because we heard our friends are using it is just as tricky as claiming the opposite, so it’s nice to finally have some actual insight (in the form of user adoption) by Wolfram.

Apple says that support for more languages will come in 2012 to Siri, and it will be interesting to see whether Wolfram will be supported internationally by then.

Permalink

Why Apple’s A5 Is So Big, And Siri Is 4S-only→

Why Apple’s A5 Is So Big, And Siri Is 4S-only

Stephen Shankland at CNET reports on a possible explanation as to why Apple decided to make Siri an iPhone 4S exclusive:

Apple’s A5 processor includes noise-reduction circuitry licensed from a start-up called Audience, and a chip analyst believes that fact resolves an iPhone 4S mystery and explains why the iPhone 4 lacks the Siri voice-control system.

Audience revealed details of its Apple partnership in January, when it filed paperwork for an initial public offering (IPO) of stock. Teardown work from iFixit and Chipworks revealed a dedicated Audience chip in the iPhone 4, but the iPhone 4S integrates Audience’s “EarSmart” technology directly into the A5 processor, the company’s S-1 filing said.

Audience confirmed in its filing that Apple signed an agreement to pay royalties for the intellectual property used in the A5 processor in the “three months ended December 31, 2011”, suggesting that only the iPhone 4S version of the A5 (the A5 CPU is also used on the iPad 2) supports Audience’s technology. Audience also said that Apple is a licensee for the next-generation IP, although they won’t be contractually forced to use it; iPhone 4-era technology from Audience also has advanced noise-reduction, but it was only with the second-generation product (the one on the A5) that Audience perfected recognition/reduction at arm’s length – thus allowing for features like voice control through Siri.

Integration of Audience’s EarSmart on the iPhone 4S’ A5 could explain why Apple decided to not offer Siri on older devices, and why it will likely only feature the voice-based assistant on newer devices going forward. A number of unofficial hacks to enable Siri on older devices surfaced in the past months, albeit with mixed results.

Permalink

Hacker Builds Custom Siri Proxy Server

A hacker known as plamoni created a Siri proxy server that could allow anyone to use it and make Siri work with a wide range of non-Apple devices. Applidium, a development firm, hacked the Siri security protocol and has explained the process so anyone can use it.

One implementation of Siri + the proxy server is sending commands to any standard thermostat with Wi-Fi capabilities. plamoni taught Siri (no jailbreak required) to send commands over the network and if you’re interested, the source code is available for free online. Anyone with an iPhone 4S unique identifier and knowledge of networking can get it working. In order to set up Siri to control your home’s temperature involves a DNS server that uses a proxy to send requests to Siri’s servers.

Video after the break. Read more

OmniFocus for iPhone Adds “iCloud Capture” To Import iOS Reminders

With the latest update to OmniFocus for iPhone, The Omni Group has figured out a way to let the app interact with iOS reminders and iCloud, and automatically fetch reminders created through the iPhone 4S’ Siri or Reminders app. Whilst you still can’t ask Siri to create new tasks directly into OmniFocus, the solution implemented by The Omni Group is very clever, and it gives you the illusion of native integration with the assistant, using iCloud and your Apple ID as a bridge between Siri’s reminder functionality, and OmniFocus.

The Omni Group has implemented a new feature called iCloud Capture that is capable of taking Reminders from your iCloud account, and move them into OmniFocus’ inbox. iCloud Capture, as explained in the Help section, literally takes input that would otherwise stay into Reminders, and moves “as much data as it can” away from Reminders and into OmniFocus. This is done in two ways: first, you have to create reminders that go into a “Reminders” or “OmniFocus” list. Second, you’ll have to log in with your iCloud account in OmniFocus’ settings, and enable iCloud Capture. Once enabled, OmniFocus will check for new reminders every time upon launch and move them into the inbox. The “upon launch” part is important, as OmniFocus can’t process iCloud data in the background – you won’t be able to add reminders via Siri, and expect OmniFocus to import & sync in the background. You’ll have to open OmniFocus for iCloud Capture to work, but it’s a minimal effort compared to the added convenience of being able to let OmniFocus seamlessly process and import reminders.

Please note that reminders will be deleted from Reminders.app as they’re moved to OmniFocus; title and dates will be preserved in the transition, but The Omni Group says that location reminders – which both Siri and Apple’s Reminders can use out of the box – will be “most effective” when they match Reminders’ contact locations and OmniFocus’ own implementation, which is quite different from Apple as it’s got more options. From my understanding of this limitation, it appears getting location reminders into OmniFocus from iCloud might work best if OmniFocus is using the same contact information rather than its “current location” or “exact location”, and I’ll need to investigate this further. For now, I can say that one location-based reminder did go through iCloud Capture, but another did not, and stayed inside Reminders.

The Omni Group didn’t add direct Siri integration, but they came up with one of the most clever ways I’ve seen to use Siri and iCloud to create reminders and have them automatically moved to OmniFocus every time you open the app. The system isn’t perfect, but has worked very well for me with simple reminders. Another good point of iCloud Capture is that Windows users will be able to move Outlook tasks to OmniFocus when they launch the app, as iCloud Capture checks for all reminders in your iCloud account, not just those created via Siri. Overall, I’d recommend giving iCloud Capture a try, and see if reminders might work better for you inside OmniFocus rather than Apple’s Reminders.

Update: The Omni Group has posted a series of videos showing how iCloud Capture works. Check them out after the break.

Read more