First look at speech recognition in iOS 5

Apple is apparently adding support for speech-to-text input to iOS 5. While speech recognition wasn’t one of the features Apple mentioned when introducing the next-generation mobile operating system, the folks at 9 to 5 Mac found a special keyboard hidden away in the latest iOS 5 beta with a microphone next to the space bar.

When you tap the microphone icon, a new screen opens showing a larger microphone. The feature isn’t fully functional yet, but it looks like you’ll be able to talk to your device for a period of time and then iOS will attempt to convert your words into speech, allowing you to enter text much more quickly — or in situations where tapping on the keyboard might not be an option.

Apple didn’t make the new feature easy to hide. The special keyboard and the microphone button are both labeled with misleading names in the latest beta. But it certainly eems like Apple is hoping to have the system in place by the time iOS 5 is released to the public.

Google Android already has speech recognition built into the operating system.

Windows Phone 7.5 to get turn-by-turn directions, speech to text, more

Microsoft has unveiled new details about the next big update for Windows Phone 7. When Windows Phone 7.5 (code-named Mango) is released later this year, it will include new Live Tiles features, better multitasking and an improved web browsing experience. But we already knew that. Now thanks to a podcast interview with Microsoft’s Brandon Watson, we also know that Windows Phone 7.5 will bring:

  • Turn by turn driving navigation
  • Speech to text support in the SMS application
  • Bing Vision — a Google Goggles-like service that will let you search the web by taking a picture of album art, bar codes, and other items
  • Bing Audio — A Shazam clone which will let you identify a song by recording a few bars with your phone’s mic

There are also reports making the rounds this morning that the Mango update will add Bing Image Search, the addition of Windows Live Messenger to the People hub, and better Skydrive synchronization in the Office hub. There may also be support for East Asian languages in the keyboard, as well as handwriting and dictionary support.

via LiveSide

 

Coming soon: Yell at your Android phone to snooze an alarm with WakeVoice

Ever wish you could hit the snooze button on an alarm without putting on your glasses, looking for your clock or phone and pressing a button? WakeVoice is a new Android app that makes it almost too easy to hit stop or snooze, since you’ll be able to stop your alarm by talking to it.

The app uses speech to text and text to speech. In other words, not only can you say ‘stop” to turn off an alarm, but you can also program the app to speak to you when the alarm goes off. Instead of listening to an annoying ringing sound, you can listen to an annoying computerized voice telling you the latest news, weather forecast, or a custom message.

The developer plans to add the ability to wake up to an MP3 playlist, radio, or other sounds.

WakeVoice isn’t available in the Android Market yet, but it should be coming soon. It will be available in two versions. The full version will run 1.49€ (about $1.96), while WakeVoice Lite will offer all the features — but you’ll have a limited number of uses.

via xda-developers

Edwin: Android app you can talk to… and it will answer

Google Android has built-in speech recognition, which means you can translate speech to text in many apps. But Edwin is one of the few that will talk back. For instance, you can ask Edwin the time, and it will tell you. Ask it for tomorrow’s forecast and a computerized voice will read it to you. You can even ask it to translate words or sentences from one language to another or to define or spell words for you.

Edwin also works as an app launcher and controller for your phone’s hardware. For instance, you can tell it to put your phone in airplane mode or turn the Bluetooth on or off. Or you can tell it to open the camera application.

Like most speech recognition apps, Edwin isn’t perfect. Or maybe it’s just my voice that isn’t perfect. Sometimes it misunderstands commands, and while you’re supposed to be able to open applications by saying “launch X” or “open X,” I’ve found that it works a lot better with “open.” But overall, it’s a pretty nifty little application, and best of all it’s available as a free download.

After the break is a video showing how Edwin responds when I ask it for the time, temperature, nearest movie theater, and other information.

via Gizmodo

Read the rest of this entry »