I tested the new Listen functionality in Pocket and it works quite well. On my Galaxy Nexus phone with Android 4.1, there was no configuration involved, nor any speech engines needed to install. I simply updated Pocket, went to a recently saved article in the app and hit the Listen button in the app menu. You can pause the reading and control the speed through a Fast/Slow toggle bar.

One of the main reasons I use Pocket as my web clipping platforms is the widespread support across many devices: I can save articles from any browser and read them on practically any device, including those running iOS. Unfortunately, the Pocket team has no immediate plans to bring the Listen feature to iOS because “iOS doesn’t currently provide a text-to-speech API.”

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• Is Android broken and if so, will Google fix it?
• A look back at mobile in Q1

Proloquo2go is just one app made by Amsterdam-based AssistiveWare, which makes software that helps those with special needs use technology. AssistiveWare worked with the linguists at Acapela, a European text-to-speech company, for a year and half on crafting these children’s voices for its mobile app. There are now four children’s voices total — two American English and two British English.

The task of creating speech recordings is enormously expensive and time-consuming, so it’s no wonder many companies have relied on adult voices digitized to sound more child-like. As founder David Niemeijer told me in an interview on Monday, “People were managing with adult voices, no one thought it was important to make the investment” in adding voices that sound like a kid would sound.

But it’s more important than you’d think for children — and their parents — to communicate with a recorded voice that sounds closer to their natural voice. Especially for very young kids who need the assistance, they can start to do speaking and voice training with the app so they can, if possible, begin to develop their own voice earlier in life, Niemeijer said. That’s why each voice can even be customized to be higher or lower.

Proloquo2Go actually began as Proloquo for the Mac (pro lo quo means “speak out loud” in Latin), and was translated to a mobile app for iOS in 2009. Then it was used on the iPod touch and iPhone. By the time the iPad came around in early 2010, the corresponding app was available the day the tablet launched. But the iPad has become by far the most popular way to use the software, Niemeijer told me. Ninety percent of Proloquo2go users use the iPad app, while 20 percent also use an iPhone or iPod touch.

There are several augmentive and alternative communication (AAC) apps available for the iPad, including Verbally, Alexicom AAC and Tap to Talk, which range from free to $99 yearly subscriptions. The iPad has come to be prized in the accessibility and special needs communities because of its portability, screen size and cost. Dedicated communication aids used to cost multiple thousands of dollars and were often heavy. Those two characteristics put them out of many people’s price range, and the bulkier form factor wasn’t suitable for really young children who needed help communicating. While Proloquo2go costs $189.99, even combined with the cheapest iPad (the $399 iPad 2), it comes out cheaper than AAC technologies that were the norm before iOS.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• Takeaways from connected consumer’s second quarter
• The evolution of consumer-media cloud storage

With iOS 5, Apple is introducing a feature that many Android users have cited as an advantage of Google’s mobile OS: improved notifications. But as our own resident Android expert Kevin Tofel has admitted, the iOS implementation of notifications actually improves upon the Google original, as well as borrowing some of its core strengths. Apple’s text-to-speech implementation could leapfrog Google’s in a similar fashion.

Nuance, the company reportedly powering the iOS speech-recognition software, is an industry leader in the tech, approached only by Google, in fact. Microsoft’s TellMe feature in its upcoming Windows Phone 7 Mango update also looks promising, indicating that this could be a key area of contention among mobile platforms. If Apple makes system-wide text-to-speech a feature accessible not only to its own iOS software but also to third-party developers via a public API, it could make Android’s voice-powered features look like an early beta by comparison. This is not only because of Nuance’s industry-leading speech-recognition software but also because in theory, any app in the App Store that wants to implement voice support might be able to include it with just a few simple lines of code.

Apple has been reported to be in talks with Nuance for quite a while now, with speculation about what a partnership between the two companies could mean, ranging from speech-recognition tech being somehow integrated into Apple’s North Carolina data center to talk of text-to-speech baked into iOS. It looks like the latter will be arriving soon, possibly supported by assistance from cloud processing provided by the former. Although keep in mind that as an unannounced feature appearing only in a prerelease build, it may be that users won’t get their hands on it until a later iOS update, or ever, if any agreement that may exist between Apple and Nuance falls through.

Apple also acquired voice-recognition-based digital assistant app Siri last year. Siri combines various data sources to allow it to provide comprehensive and convenient answers to a question asked by a user. In theory, Apple could use it in combination with Nuance tech to create an intelligent Swiss army knife voice-activated launcher/search/command interface for iOS. A system-wide voice-input button like the one spotted by 9to5Mac in this beta build makes that a much more realistic possibility.

Some will accuse Apple of “copying” Android’s best features in order to stay in the mobile race, but I say as long as in doing so Cupertino makes the iPhone user experience even better, more power to it.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Siri: Say hello to the coming “invisible interface”
• Mobile Q2: Smartphone growth surges; iPad’s rule continues
• Why Apple Could — and Should — Bring Voice Recognition Technology to our Phones

I took the software for a spin, and it works as advertised, but there’s definitely room for improvement. The setup process was fairly simple: Download, install and choose which online providers you want the app to monitor. For now, RadioMe supports Twitter, Facebook, Gmail, RSS feeds and text messages, although I couldn’t get the software to link up with Facebook. I did connect my Twitter account without any problems, and as the tweets began flowing, so too, did the audio updates in a computerized voice.

I’m not too concerned about the voice because it handled most pronunciation quite well. And if you don’t like the default voice, you can purchase and install other voice engines from the Android Market. The bigger issue to me is two-fold. First, as the app works now, it simply reads every update that hit my Twitter feed. I see no way to configure the software to only read mentions or replies, for example. And second, a huge portion of social networking is the shared links and photos that people provide. Each time RadioMe sees one of these in a tweet, it simply says “link.”

There are other welcome configuration controls, however. Since the app runs in the background, the developers accounted for music listeners that still want to hear social networking updates: You can set an audio reduction level for music volume to decrease while updates are read. The number and frequency of updates can be modified, as well.

I like the concept of RadioMe, which is similar to Vlingo and other text-to-speech solutions. The folks behind the app are going after a social networking radio station, which is an interesting idea. But for some, the software would be far more useful, at least for Twitter and Facebook, if it offered more control over which social networking updates it reads instead of simply reading them all. Since it’s a beta product, there’s room and time for improvement. Once a finer level of control is provided, RadioMe could offer more appeal by speaking less, or at least offering that as an option.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Analyzing the wearable computing market
• 4 iPad apps to help wrangle data
• Facebook’s IPO filing: ideas and implications

So what does that actually mean? The site presents a montage of images, video, animations and other visual resources culled from search results, and overlays it with real-time narration, using text-to-speech technology. The result is information as a watchable experience.

For me, the actual experience of watching the content is eerie and unsettling. I’ve never liked the way text-to-speech technology sounds; it’s strange and unnerving to my ears and brain, sounding almost, but not quite, human in a way that could mean trouble, like HAL in 2001: A Space Odyssey.

My first search was for “Tok,” the rural Alaskan community where I live and work. The text-to-speech voice sounded nearly human, but with mispronunciations and odd inflections that are inherent in the technology. Qwiki pronounced my community as “Tawk” rather than the correct pronunciation, “Toke.” Right away, this interfered with my ability to appreciate the visual montage.

The audio was accompanied by a flowing stream of imagery that seemed at first to correspond with the narration. But on closer examination, many images had nothing to do with what was being said, other than being from Tok. For example, when the narration mentioned Tok School, the images that appeared were of a gift shop and an RV park. Another image of a burned-out old gas station showed up, and I immediately wondered how I could remove such a photo, and replace it with something more representative of the community. Ditto for an image of a coffee shack that isn’t even in Tok, but is over 200 miles away, near the community of Glennallen.

My second search was for “karaoke.” I found the experience of this information less off-putting. However, the images in the presentation, while colorful and interesting, seemed less familiar. Many turned out to be of displays and equipment common in Japan.

My third search was for “social media.” The narration was a bit convoluted and the visual presentation consisted of a single static screenshot of Flickr.

My final test search was for…me. Here’s what I found (login required). And here’s what it looked like–apparently I have one of the dirtiest minds in business–or at least that’s what was displayed during the entire presentation!

Qwiki may give us a new phrase to replace “Googling ourselves.” We can now “get a Qwiki” and “give a Qwiki.”

Currently, Qwiki covers over two million reference terms, which feels sufficient for pretty good results when searching for a not-too-uncommon term. You can also search for people and places.

For me, the visual and audio dissonance of Qwiki was initially disturbing, but I tried to look past that and appreciate that I was “experiencing information.” Presentations are short–about 30 seconds–which is just enough time for narrative content from a paragraph or two from a Wikipedia entry.

Conceptually, Qwiki is a fascinating step forward in the presentation and consumption of search results. Since the site is in its alpha phase, it’s available only by invitation, although you can request one at Qwiki’s website. You’ll probably start itching to fix what isn’t working, but Qwiki doesn’t yet have a Wikipedia-like system for collaborating on editing information. However, the company is very open to input on ways to improve the experience.

Give Qwiki a try, and let me know what you think of it. What implications do you think it will have for the future of search?

Related content from GigaOM Pro (sub. req.):

• Social Media in the Enterprise
• Email: The Reports of My Death are Greatly Exaggerated
• HTML5’s a Game-Changer for Web Apps