Even though Microsoft and ARM have a relationship that dates back to the mid-1990s, Microsoft Windows is, for good reason, most closely associated with Intel (WinTel being the predominant desktop combo in enterprise settings). Yet Ed Sperling gets it exactly right when he points out that ARM performs better than the Intel ATOM processor when it comes to power management, in battery-powered mobile or embedded devices.

Taking the two announcements together, we can expect Windows Phone-based products to have sufficient battery life and do a better job of recognizing and rending spoken utterances than current models.

Keeping with the Hawaiian theme, the program incorporates initiatives that have been conducted as part of Microsoft’s “Mobile Assistance Using Infrastructure” (MAUI) Project. Its purpose is to enable “a new class of cpu- and data-intensive applications that seamlessly augment the cognitive abilities of users by exploiting speech recognition, NLP, vision, machine learning, and augmented reality”. Based on experience with WinMobile 6 and 6.5, it determined that such a project is required to define how to overcome issues surrounding battery life (which it calls “energy limitations of handhelds) “by leaveraging nearby computing infrastructure”, which is believed to encompass resources in Microsoft’s storage and compute cloud (Azure) accessible over wireless networks (WiFi, WLAN, FemtoCell…)

In a nutshell, Microsoft says that the mobile “platform” runs on a Windows Mobile 6.5 smartphone (with Windows Phone 7 “in planning”). The services that are invoked in Microsoft’s cloud include Bing Maps for mapping services, and Windows Live ID for user identification. More detail on MAUI can be found on the Microsoft Research Web site.

Yesterday the “Inside Facebook” blog featured this post about a service called VoxCard, which enables the denizens of Facebook to festoon their friends’ or their own walls with talking postcards. They start by adding the VoxCard application to their Facebook account. Then, albeit en Francais, they can choose a character from a gallery that includes of celebrity caricatures, popular icons or quirky personalities. They can then choose and customize a standards message or type in an original message from scratch.

Messages can be posted to the originators “wall” or to the wall of a friend. As alternatives they can be embedded in messages to any number of people in the Facebook member’s contact list. When the recipient clicks on the card, it recites the message in the character’s voice (with a French accent bien sur). According to the Inside Facebook post, there app is seeing about 200,000 “active monthly users,” a small fraction of the hundreds of millions of people on the popular social networking platform. Still, it is there for all to see, hear and play around with.

L4 Systems’s Connected TV is a bit different. It is not yet available for all to see, hear and play around with. Instead, the company will debut its technology in the fall, to coincide with the introduction of Google TV, Tivo Premiere and a forthcoming Yahoo! TV (which I have not seen mention of elsewhere). The Seattle area software company has determined that it is not too early to debut and demo its flagship software product which aims to deliver “an always-available visual and speech recognition user interface to interact with features in the living room and on the go.”

It plans a simultaneous release of two applications. L4 Connected TV and L4 Connected TV Mobile. Together they turn an iOS phone, Android or Blackberry into a voice navigation module for “Connected TV” products, including “Yahoo Connected TV” running on on Samsung, Sony, Vizio or LG sets; Google TV and TiVo. This will provide users with the ability to navigate an aggregation of content that includes social networks, local search and music. The sources, as noted in the company’s press release, will include Facebook, Twitter, Google Buzz, Blogger, Bing, and Apple iTunes or Amazon Music Store. A relatively static “demo” with screen shots can be viewed here.

The fall season promises to bring a number of speechable moments to the TV-watching public. GoogleTV includes voice control in its offering already. Mobile dictation specialist Promptu got its start as AgileTV, a company whose charter was to provide voice-based control of the cable TV program guide and program selection process. The addition of L4 to the mix of technology providers is a tacit admission that navigating all the content, features and services offered through TV screens today and the future is getting complicated and voice commands should serve as the most efficient way to cut through the clutter.

Of greater interest from our point of view is the integration of Microsoft’s Tellme service into the Windows Mobile mix – this time as the foundation for voice-based turn-by-turn driving directions. It has been over a year since Greg Sterling posted this assessment of Tellme Mobile and its ability to support voice-based search, messaging and call origination. As one of the originators of the whole “Voice Portal” concept, Tellme has had the ability to offer turn-by-turn directions for a number of years.

Starting with the first Droid and its “car mount” (introduced in October 2009), smartphone makers have had their sights set on the Personal Navigation Device (PND) market. Just last week, Nokia stepped up competition by creating a marketplace for “own voices” to provide turn-by-turn directions for its OVI Maps application. Making turn-by-turn directions part of the Bing Mobile application makes navigation an integral part of the local, mobile search experience and is a very nice touch.

Microsoft already incorporates directions (sans the “voice guide” feature) into the Bing app on iPhones and Androids. The new application is designed to work on the Sprint, T-Mobile, or AT&T networks, and could emerge as an important differentiator in what is emerging as an increasingly competitive market.

Lassa asserts that the GM/Google relationship would place emphasis on a better phone-to-car interface, as opposed to the voice control and voice user interface that Microsoft’s Speech Application Group has played up. Thus GM’s approach will enable drivers to use their phones to do such things as start or turn off their cars, lock and unlock doors, and make other adjustments. It was not spelled out explicitly in the article, but given Google’s efforts to invoke automated speech recognition whenever a keyboard comes into play on a mobile device, it is highly likely that all of these functions can be voice controlled – making starting your car another “speechable moment”.

As for the supposition that Android in the car spells the end of OnStar, that is highly unlikely. Lassa notes that turn-by-turn directions through OnStar would become unnecessary because Android phones using Google Maps and a special mount have been successfully deployed for in-car navigation. But OnStar has been sold more as a safety feature and remote diagnostic service. The Android operating system in the car is more likely to augment, rather than compete with OnStar.

The prospects for more automobile-based Android apps is provocative. The car is destined to be the most fertile spawning ground for speech-based apps and the prospects for Android-oriented developers to define a range of “hands-on-the-wheel/eyes-facing-forward” capabilities and activities is very promising. Meanwhile, Ford remains ahead of the game with a well-defined, and now time tested, suite of voice control applications for frequent activities like carrying out phone conversations, messaging and controlling the car’s entertainment system.

Nokia purchased the leading provider of digital maps, NavteQ, in October 2007 for more than $8 billion. It was a clear statement that it was prepared to position its smartphones as a more functional and affordable alternative to single-purpose personal navigation devices (PNDs) from TomTom, Garmin or Magellan, bearing in mind that these were companies that in many cases were customers of NavteQ and used its files to support their PNDs.

When Google and Verizon introduced an automobile mount for the Droid in late 2009, it not only intensified the belief that single-purpose PNDs’ days are numbered, it was a direct challenge to Nokia and its OVI Maps service. Thus Own Voice is the product of heightened competition and it will serve as a test to see if the guiding voice emanating from a navigational system (as opposed to price, appearance, accuracy, ease of programming or some other factor) can emerge as a competitive advantage.

As the Web site indicates, it takes quite a few steps to initiate use of Own Voice. At a minimum you must own one of the listed phones, which include Nokia 5800, Nokia X6, Nokia E72, Nokia N97, Nokia N97 mini and all upcoming GPS-enabled smartphones. You must also have purchased and installed the OVI Maps app from the OVI Appstore. Own Voice is a free add-on to the OVI Maps Application. At that point subscribers are navigated through the process of recording 53 discreet spoken elements, including the ever-popular “You have arrived!”. Then “Voila”, they have created a personal navigation application that employs their own voice.

In this era of immediate gratification and real-time communications, it is ambitious and probably unrealistic to think that new users will spend the hour or so it might take to record 53 utterances to their own satisfaction. Just think of how many times it took for you to get “the right greeting” on your mobile phone’s voicemail service. Now think of doing that 52 more times. It is much more likely that they will opt to choose the pre-recorded voices that other Own Voice users have “shared” in the Own Voice marketplace. It will never be as big as “Charlie Bit My Finger on YouTube, but Nokia is creating the sort of social platform that can provide an ego boost for a creative, self-identifying voice talent. There were already scores of examples in dozens of languages on the site, which includes a column of “Featured” voices, as well as the obligatory “Most Popular” page to help people identify a short list of candidates for the voice they’d like to hear when they’re looking for directions in a strange town.

Nokia deserves credit for using a voice-based service to differentiate its mobile product. Own Voice for Ovi Maps’ planners also deserves style points for creating yet another platform for someone’s creative work to “go viral” (though we’d be interested to know the size the “audience” made up of users of the latest rev of OVI maps on selected Nokia smartphones amounts to these days). It is just the sort of thing that may be daunting to casual users but challenging enough to attract a sufficient number of creative types to make for some very entertaining, and well differentiated, PND-based experiences.

Siri has already proven to be a popular application for iPhone, and has been deemed even better for the iPad where, with closer ties to the parent company’s internal labs, it would help define Apple’s tablet as the ultimate personal assistant. The founders of Siri had already been instrumental in bringing millions of dollars (and person months) invested at SRI’s Artificial Intelligence Center in the CALO/PAL project (where the word pal refers to a “Personal Assistant that Learns). It was the largest and singularly most ambitious project to apply artificial intelligence to solve everyday, recurring problems.

By buying Siri and hiring the core staff, which includes founders Adam Cheyer, Dag Kittlaus and Tom Gruber. Cheyer served as program director for CALO. Kittlaus applied his extensive experience in mobile services at Telenor, Motorola and Screen3. Gruber’s insights into the practical application of artificial intelligence has been an important factor in shaping the services of LinkedIn, SocialText and PowerSet.

There is no confirmation from either Siri or Apple that the acquisition is pending. Yet the prospects for embedding a “personal assistant that learns” across Apple’s entire “iLine” (iPod, iPad and iPhone) as well as its line of PCs and cloud-based services (meaning the iTunes Store, etc.) is enticing. When we first wrote up Siri in February we noted that it had forged relationships with Web sites that put an emphasis on “results”: Fandango for movie ticket buying or Open Table to book a table at a restaurant. The company’s revenue plan was to take a percentage of the transactions it supports. Clearly, it would take a while for a small company like Siri to build the subscriber base and transaction flow to make its revenue plans “lucrative.”

As part of Apple the company will find that it has a longer time to build up the revenue stream while, at the same time, it will be easier to scale the business using Apple’s existing resources. Clearly a win/win.

And let’s not forget that it is also a coup for Nuance Communications, which provides the voice recognition resources for Siri as well as for the “native” Voice Command apps on the iPhone. Congratulations to all.

In its press release, the company claims to be the first to offer “voice addressing” for text messages. The company claims high levels of accuracy because it is the only application to be “fully integrated with the Android SMS client and uses Smartword, Promptu’s editing tool that lets users see and select from likely word alternatives.” The application debuts today (April 26) for the Motorola Droid with features taht include: voice addressing, “full integration with the phone’s messaging service (including message exchanges threaded by recipient), thumbnail images displayed for contacts and “one-touch navigation to all recent incoming and outgoing messages.

As reported here, Promptu’s revenue model will test the value that users attach to transcription services. If it remains true to form it market bundles of “voice credits” in increments of 50 for $1.99; 150 for $3.99 or 250 for $4.99, as it does for iPhone users. This is in contrast to its closest functional rivals Vlingo (which has charged for a $17.99 for its application on the Blackberry) and Nuance Dragon Dictate, which has not yet announced a revenue model. Vlingo, by the way, has a “beta” of its application for the Android, which has been available since mid-March.

Along with Google, Microsoft and (to some extent) AT&T, these are the initiatives that are shaping the way that spoken words will be integrated into the mobile messaging flow.

Heck, Apple features access to its Software Developers Kit as prominently as its “How to Order” link. So it is no wonder that the cottage industry that is “user generated content” will be bigger for the iPad than it is for the iPhone (well, at least that’s my prediction).

But can it talk? Thus far, it has been hard to discover the full spectrum of applications that run on the iPad. Apple claims there are already 150,000 iPad Apps. Yet search for “iPad” in the iTunes App store delivers a scant 2 pages. But wait! A search of the iTunes store for “iPad apps” that feature “Speak” yields a number of applications that don’t appear when the “Select All” option is invoked. These include:

Speak4It from AT&T Interactive R&D (listed under “Lifestyles”)
Speak It! Text to Speech (listed under “Utilities”)
iSpeech for the iPad (listed under “Education”)

Thanks to this article in Gizmodo, I was clued into the fact that a version of the Dragon Dictation app that I have on my iPhone has been made available for the iPad. The availability was confirmed in this press release from Nuance, which contains a link (which opens iTunes and delivers you to the iTunes Store where the app is tagged as “Business”.

Clearly, “speech enabling” apps defy categorization. I dare say that, before I shell out $69+tax for an iPad keyboard/dock or wireless keyboard, I will try my hand at dictating email using the free version of Dragon Dictation.

The Vlingo application on iPhones, as well as Blackberries, already support spoken origination of SMS, email and status updates for Facebook or Twitter. In my experience, it has been very good at distinguishing between commands (such as “Email Ted Stevens…” or “make an appointment”) and content, thus minimizing the need to do a lot of screen poking. In its marketing literature, Vlingo calls “the reverse functionality”, and sees it as a means to accomplishing the tasks that involve rendering written words as speech.

Vlingo’s Dave Grannan has long made it clear that the company doesn’t endorse the use of mobile devices while driving. Therefore they consider this a safe, hands-free way to retrieve “must-have” messages.