Last week I downloaded Jott, a free application, from the iTunes store. It is a serviceable voice recorder, so my wish is fulfilled.

But the beauty of the third party software community concept is that motivated, talented people in hungry startups will go beyond what’s justifiable in a large company like Apple, and this is what Jott has done. It doesn’t just record voice memos, it transcribes them into written text.

It works very well. It uses people to do the transcriptions. I am not sure if the utterances are preprocessed with Automatic Speech Recognition (ASR) and transmitted to humans for verification and correction, or if it is entirely done by people in a call center somewhere. When I mumble the text comes back as “Unclear,” but I can still play back what I said and recognize it for myself.

There are a few other transcription-type applications out there. Spinvox and PhoneTag transcribe voicemail into SMS and email. A great idea. Nuance, the world leader in voice recognition technology, announced a similar service in April.

In contrast to the foregoing, Yap is 100% automated, so to avoid mistakes it has the user verify its efforts. You speak the text you want to send as an SMS (or that you want to search the web for) and Yap renders it as text on your phone’s screen. You correct it and send it off. Yap doesn’t appear to be deployed yet.

Similar to Yap, but already deployed in the real world is Vlingo. I went to the Vlingo website to download a trial, but didn’t when I discovered I would have to buy a Blackberry to try it on. Vlingo was recently adopted by Yahoo! to power its onSearch mobile product. Nuance is suing Vlingo for patent infringement. Nuance has announced an application like this for the iPhone, but a search for “Nuance” in the iTunes store doesn’t yield any results for it yet.

Another ASR granddaddy is Tellme (now owned by Microsoft), which powers the Sprint Live Search service. Tellme also lets developers do free hosted low-volume implementations of their concepts in VoiceXML.

Getting back to my iPhone wish list, I am still baffled as to why it doesn’t do cut and paste. The argument that it would require an awkward user interface was exploded a year ago.

The idea of the article is that voice over Wi-Fi for cell phones is competing with femtocells, and that femtocells may win out. The article distinguishes between business voice and consumer voice, saying that service providers see femtocells as “an important stalking horse for greater control of corporate customers. ” This gives a hint of why femtocells may be unattractive to businesses: many of them would rather not yield this control.

Consumer voice service is controlled by service providers. They have three options in this space: do nothing, deploy femtocells or deploy Wi-Fi. Do nothing is the obvious best choice, since neither of the other options carries a revenue upside. But poor coverage in a home discourages usage and risks cancellations of subscriptions. So in areas of poor coverage something like femtocells or UMA (voice over Wi-Fi) is attractive to service providers. For both technologies the service provider subsidizes the wireless router, but femtocells will remain more expensive than Wi-Fi routers because of their lower sales volumes, so Wi-Fi is more attractive on this count. But UMA requires phones with Wi-Fi, while femtocells will work with any phone in the service provider’s line-up, including legacy ones. So the customers’ experience of femtocells is better - they can choose or keep the phone they want and still get improved coverage at home. This benefit of femtocells clearly outweighs the marginal price advantage of Wi-Fi routers. Femtocells may help subscriber retention in another way: a Wi-Fi router is not tied to any particular cellular service provider, while a femtocell only works with the carrier that supplied it.

The situation in businesses is different. They generally prefer to control their own voice systems, which is why they have PBXs. But a substantial number of business calls are now made on cell phones, even on company premises. These calls don’t go through the PBX, so they are not least-cost-routed and they are not logged or managed by the IT department. Femtocells don’t fix these problems, but Voice over Wi-Fi does. Not service provider Voice over Wi-Fi, like UMA, but SIP-based Voice over Wi-Fi from companies like DiVitas and Agito. What about phone choice though? Won’t corporate customers be stuck with a limited choice of handsets? The answer is yes, only a limited number of phones have Wi-Fi: less than 10% of those sold in 2008. But in the category of enterprise smart phones, like the Nokia Eseries and Blackberries, the attach rate of Wi-Fi will soon be close to 100%.

So femtocells are a good way for service providers to remedy churn caused by poor residential coverage for consumers, but Wi-Fi may be the better option for businesses that want to regain control over their voice traffic.

Ozmo’s pitch is that their special version of Wi-Fi is faster, less prone to interference, cheaper and more power efficient than Bluetooth. Though slow for Wi-Fi, Ozmo’s data rate of 9 Mbps is way better than the 3 Mbps of Bluetooth 2.1 plus EDR. Concerning interference, Bluetooth transmissions are in the same 2.4 GHz frequency range as Wi-Fi, so careful engineering is always needed for coexistence; but coexistence is not an issue for Ozmo, because it is a flavor of Wi-Fi. The lower-cost part of the story is that while an Ozmo chip for a peripheral is roughly the same price as a Bluetooth chip, it saves the need for a Bluetooth chip in the host device. So if Ozmo’s claims of double the battery life of Bluetooth turn out to be valid when tested, the pitch holds water.

Bluetooth is unshakably incumbent in the billion-unit-per-year mobile phone market; most phones are now shipping with Bluetooth. But in laptops the Bluetooth attach rate is still well under 50%, while the Wi-Fi attach rate is close to 100%. So it makes sense that Ozmo is focusing initially on the computer market. Ozmo enables a laptop to use wireless peripherals like mice, keyboards, game controllers and headsets with a software modification to the Wi-Fi rather than having to include a USB wireless receiver in the package.

Ozmo is also focusing on the Consumer Electronics market, where Wi-Fi is gaining traction, and where the ability to support wireless remotes and similar peripherals can be added ‘free of charge’ to a device that already has Wi-Fi.

Since this is a compelling proposition, OEMs in these markets are likely to fall into line relatively easily, and indeed Ozmo has already secured the support of the biggest fish in the PC pond, Intel.

It may be a harder sell to the peripheral manufacturers. For them it’s not just a software upgrade. It is a new product line, one that depends on a single source for a key component and that sells into a currently non-existent base of host devices.

On the other hand, the peripheral device manufacturers may not be oposed to a new product line - they already have a plethora of products, using a variety of connectivity technologies including proprietary ones. Adding another one may not be too arduous, especially if it removes the need for a USB part. Ozmo could overcome the non-existent base issue by supplying, or convincing Intel and the other PC Wi-Fi chip vendors to supply, upgraded drivers for legacy Wi-Fi devices.

Why is this huge news? Well, as the press release points out, by 2011 annual shipments of cell phones with Wi-Fi will be running at roughly 300 million units. The Wi-Fi in these phones will be used for Internet browsing, for syncing photos and music with PCs, and for cheap or free voice calls.

The certification requirements for Voice-Personal are not aggressive: only four simultaneous voice calls in the presence of data traffic, with a latency of less than 50 milliseconds and a maximum jitter of less than 50 milliseconds. These numbers will produce an acceptable call under most conditions, but a network round-trip delay of 300 ms is generally considered to approach the limit of acceptability, and with a Wi-Fi hop at each end running at the limit of these specifications there would be no room in the latency budget for any additional delays in the voice path. The packet loss requirement, 1% with no burst losses, is a very good number considering that modern voice codecs from companies like GIPS can yield excellent sound quality in the presence of much higher packet loss. This number is hard to achieve in the real world, as phones encounter microwave ovens, move through spots of poor coverage and transition between access points.

Since this certification is termed “Voice-Personal,” four active calls per access point is acceptable; a residence is unlikely to need more than that. Three of the four access points submitted for this certification are enterprise access points. They should be able to handle many more calls, and probably can. The Wi-Fi Alliance is planning a “Voice-Enterprise” certification for 2009.

There are several things that are good about this certification. First, the WFA has seen fit to highlight voice as a primary use for Wi-Fi, and has set a performance baseline. Second, this certification requires some other certifications as well, like WMM power save and WMM QoS. So far in 2008, of 99 residential access points certified only 6 support WMM power save, and of 52 enterprise access points only 13 support WMM power save. One of the biggest criticisms of Wi-Fi in handsets is that it draws too much power. WMM power save yields radical improvements in battery life - better than doubling talk time and increasing standby time by over 30%, according to numbers in the WFA promotional materials.

The conventional solution in these cases is to upgrade the Ethernet network while junking the old phone wiring.

Phybridge proposes to leave the Ethernet network alone, and to reuse the old phone wiring to implement a parallel data network, using Ethernet over a flavor of DSL. This is similar to HomePNA, but aimed at business use rather than consumer, and done point-to-point rather than into a shared medium.

The solution consists of two parts: a central box called “Uniphyer” has 24 ports connected to the legacy phone wiring. At the other end of each cable run is a “phy adapter” the size of a pack of cigarettes that you plug into the legacy phone jack, and into which you plug your Ethernet VoIP phone.

The Uniphyer provides power over the same copper pair as the data, so you can plug power-over-Ethernet phones into the client adapters.

The data rate is 3 megabits per second upstream, 30 down. This is slow for a data network, but certainly adequate for VoIP, so an organization that is replacing a conventional PBX phone system with a VoIP one may find Phybridge a cost effective solution if their existing data network isn’t up to VoIP, and the required improvements are extensive.

The Uniphyer is scheduled to launch at the end of September.

The answer is yes, but when you delve into the reasons why you may discover that an 802.11n handset can still have comparable, or better battery life than an 802.11g one.

The big power drain for 802.11n is MIMO, for two reasons. First, MIMO demands a separate radio transmitter for each of its channels. In the Farpoint white paper linked above, testing was done with six transmitters - 3 at 2.4 GHz, 3 at 5GHz. The 11n specification allows up to 4 MIMO channels, and Wi-Fi certification requires at least two. Each of these transmitters burns as much power as the single (or dual in the case of an a/g AP) transmitter in an 11a or 11g access point. A second increase in power demand by 11n comes from the increased processing load not just because of the increased number of channels, and not just because of the increased data throughput, but also because each individual MIMO stream places a heavier processing load than a single 11a or 11g stream.

But the Wi-Fi Alliance (WFA) has waived the MIMO requirements for handsets, allowing 802.11n certification for single-radio devices. So none of these increases in power dissipation needs to apply to handsets.

Single-channel 802.11n still requires more processing than single channel 802.11g, because of advanced features like STBC and LDPC, but STBC and LDPC are amenable to hardware implementation (which reduces their power demand), and these and other advanced features of 802.11n improve “rate at range,” meaning that the transmitter is active for shorter times, and can transmit at lower power.

The net is that Redpine Signals, a pioneer of 11n for handsets, claims that a handset using the Redpine 11n chip actually has better battery life than it would with a competitor’s 11g chip.

Wi-Fi state of the art is a rapidly moving target, and over the past 12 months there have been startling improvements in power efficiency. I have written here about the new Atheros chip, for example. So if the latest 11g handset chips are more power efficient than 11n competitors, it is more a function of their recency than their adherence to 11g.

The benefit of 5 GHz operation is compelling for Voice over Wi-Fi, and it will be hard for handset vendors to promote the decade-old 802.11a over 802.11n. 802.11n is already the Wi-Fi flavor of choice for access points and PC clients, and it soon will be for handsets, too. How soon? It’s hard to say. So far the only chip vendors to announce 11n for handsets are TI, Redpine Signals and Conexant, and Conexant exited the handset Wi-Fi business just two months after it announced this chip. No phone is yet shipping with 802.11n although TI said it was sampling its WiLink 6.0 with 11n in February 2007. The Wi-Fi alliance has not yet published its Handheld profile for 802.11n certification. On the other hand, ABI research in September 2006 predicted that the majority of the 300 million Wi-Fi enabled handsets to ship in 2011 will support 802.11n.

If 802.11n handset shipments fall short of this prediction, it won’t be because of battery life considerations.

You may remember that last year Bluetooth subsumed WiBree as an ultra low power consumption variant. Now something analogous is happening in Wi-Fi, where a startup called Ozmo has come up with an ultra low power variation of Wi-Fi, which it is pitching as a substitute for Bluetooth.

Ozmo is not the first to propose this. Nanoradio claims that its implementation of Wi-Fi consumes way less power than Bluetooth, and they have a reference design for a Wi-Fi headset to demo the concept. Atheros has also recently announced a low-power Wi-Fi chip - so low that they claim “near zero” stand-by power consumption.

So Ozmo may not be so radical in its low-power claims, but it also claims low-cost. The Ozmo chip is priced the same as Bluetooth chips, which Ozmo estimates are between 1/4 and 1/3 the cost of Wi-Fi chips.

The Atheros and Nanoradio chips are full implementations of Wi-Fi; Ozmo’s is, according to this EETimes article, a “stripped down” version. This may mean that it is a non-standard version, in which case it would need a new certification program from the Wi-Fi Alliance - not a major obstacle, I presume, in view of Intel’s strong support for Ozmo.

So why would anybody want a version of Wi-Fi to substitute for Bluetooth, when we already have Bluetooth? Well, while most phones come with Bluetooth these days, in the PC world Wi-Fi is far more common than Bluetooth. So for peripherals designed exclusively for PCs, like mice and keyboards, Wi-Fi makes more sense than Bluetooth simply on the grounds of out-of-the-box compatibility. Add the technical benefits, like coexistence (Bluetooth and Wi-Fi tend to step on each others’ toes) and superior performance (Ozmo claims double the battery life in similar applications to Bluetooth), and it seems like a no-brainer.

The factors that could cause it to fail are ease of use and price. The setup of the peripherals with the PC must be completely automatic, and the operation of the devices must be flawless. Peripheral manufacturers must produce Wi-Fi versions of all their wireless products, and price them the same as their Bluetooth versions. The Wi-Fi drivers on the PC side will have to be modified in order to permit the PC Wi-Fi to communicate simultaneously with access points and peripherals. This is being done anyway by Intel, in a project called “Cliffside.”

All these impediments are mitigated considerably by Intel’s involvement. Intel makes most of the Wi-Fi chips in PCs, and presumably these chips will support the Ozmo products. With this assurance, peripheral manufacturers will feel comfortable going ahead with Ozmo-powered products.

One of Agito’s unique technical claims is that it employs location based technology as one of the factors to determine the handover between Wi-Fi and Cellular (and vice versa).

The RoamAnywhere Mobility Router adds the element of location awareness to solve the challenging problem of routing between multiple networks.

From the Agito website

The Mobile Intelligent Roaming component of Cisco’s new Mobility Services Engine (MSE) also uses location as one of the factors (the other are signal strength and signal quality) to determine when to hand off between networks. This incorporation of location awareness into the MSE’s handover recommendation appears to weaken Agito’s claim to uniqueness in this respect, so why is Agito so enthusiastic about the MSE?

The MSE boils down an assortment of metrics related to handover decision-making to a simple binary message – link-up or link-down – which it sends to a third party eFMC controller like the Agito RoamAnywhere Mobility Router. The message is a recommendation, not a command, but since the MSE sends just this binary event, rather than any of the metrics that go into it, Cisco seems to be claiming responsibility for the decision about when to hand over.

Rather than simply obeying the link-up or link-down event by changing networks, the Agito mobility router takes these events as triggers to run its own handover-decision algorithm, using metrics gathered by the handset. This makes sense for several reasons. First, the Agito location determination mechanism is more precise for this particular application, since it knows when it is going through a doorway, whereas the Cisco location mechanism has a more general idea about the perimeter of the building. Second, the entire Agito handover-decision algorithm has to be maintained regardless of how good the MSE is at making the decision, because in many places there is no MSE (at home, in public areas, in buildings with non-Cisco networks and in buildings with Cisco networks lacking an MSE).

So Agito views the information from the MSE as a mere additional factor for its handover decision-making, rather than a substitute. On the other hand, Cisco’s marketing program around the MSE is a huge benefit to Agito.

Even though this part of the MSE will not be released until later this year, Agito’s Mobility Router works fine without it; the Cisco Motion announcement has led to a burst of interest in Agito from Cisco’s channel partners. According to Cisco, it has 65% share of enterprise Wi-Fi infrastructure, so Pej Roshan, Agito’s V.P. of marketing, anticipates that the relationship will “slingshot us to the next level of sales and customers.”

So I was pleased to hear an aside from a Cisco executive in the webinar about the new Cisco Mobility vision. A participant asked Brett Galloway if delays in the ratification of 802.11n are holding back deployment. He responded:

We don’t see it. Technology adoption is a cycle. Late adopters are more conservative. Large numbers of 11n clients are shipping. The driver is Wi-Fi Alliance Draft N certification. There is comprehensive agreement around the ecosystem for interoperability testing using the WFA testbed. Cisco announced recently that it had shipped 50,000 11n access points.

The critical point here is that a final 802.11n specification is not needed for successful deployment. The main thing you want from a Wi-Fi device is that it is interoperable with all the others, and that’s what the Wi-Fi Alliance Draft N certification delivers - at least to the same extent that 11g devices are interoperable.

While 50,000 is a trivial number in the context of Wi-Fi access points, remember this is just enterprise-grade APs, and doesn’t include anything from Cisco’s Linksys division. When the litigation goes away and the final specification is published, all the access points out there are likely to be upgradable with a download. So go ahead and enjoy the benefits of 11n like so many other people are doing.

As you know, the Atom is Intel’s prospective ARM-killer, but Intel will have some very heavy sledding to do to achieve this goal. The first obstacle is size and complexity. The Intel Architecture instruction set is much more complicated than the ARM, with decades of legacy dross gunking it up. The core proposition of IA for handsets is that IA is compatible with all the PC software and websites out there, so none of the legacy stuff can be left out. Intel has two factors working in its favor for this issue: first, a lot of the little-used functionality of IA can be implemented in microcode, which takes minimal die space. Second, Intel leads the world in process technology, and as it moves to the 32 nm and below nodes, vast numbers of transistors can be fitted into tiny chips. This shrinkage, and advances in power management technology, may enable the next generation of Atom to get into the same ballpark as ARM for battery life.

The second obstacle is incumbency. There are vastly more ARM processors out running in the world than there are Intel Architecture processors. The ARM is the processor used by effectively all phones and smart phones, and there are even more ARMs in other embedded applications. While Intel can point to a huge independent software developer community for IA, with a rich array of third party developer support, so can ARM for its architecture. And it’s not clear that the Atom will have a significant performance advantage over the ARM. ARM is not standing still on its architecture, delivering faster, lower power chips in step with Moore’s law just like Intel.

The incumbency issue also has a political aspect. Handset manufacturers are disinclined to hand Intel the kind of architectural franchise that it has in the PC world, so the Atom will have to have a compelling, essential advantage over the ARM in order to displace it in any but niche applications. Intel hopes that “the real Internet” is that advantage, but the iPhone has demonstrated that the real Internet can run very nicely on an ARM-based phone. In any case, within three years more smart phones will be shipping than PCs, and the other handset manufacturers will probably have caught up with Apple on the browser front, so it is possible that more Internet browsing will be being done with ARMs than IA chips.

The third obstacle is business model. To satisfy the stock market, Intel requires aggregate gross margins around 60%. This will be very tough to achieve in the mass-market for handset processors, because the margins on competing ARMs are much lower.

Good background on the Atom can be found in this EETimes article, this Beyond3d article, and this one by my colleague Linley Gwenapp.