I've recently tried out 24 different Voice Applications from Windows Phone 8 market place. There's a very brief summary of the App and my opinion of the quality. Also I've noted the provider of the speech technology, but note that in most cases the UI and experience are the responsibility of the app developer not the speech technology provider. There are many worthy efforts here while some simply didn't work at all when I tried them. Take a look yourself and decide what you like. A noteworthy capability that Windows Phone 8 has introduced is the concept of deep linking. Although only a handful if the apps take advantage of this it gives third parties the opportunity to make a direct link to their application by registering a keyword with Windows Phone that allows you access the app functionality as though it were native to Windows Phone. This is way cool because when properly used it creates a natural connection between the brand name and the capabilities of the application and the brand.
I've divided applications into into 6 broad categories, Command and Control, Dictation, Games, General, Search and Translation. Note that I didn't specifically review the built in voice capabilities of Windows Phone 8 that include Voice Activated Dialing, Voice to SMS, E-mail composition and Application launching as well as the speech API used by eight of these apps.
Command and Control: Theseapplications do something for the user. Each of these apps uses the Win-RT speech API. Their biggest weakness is that their grammars are not robust and the users may have difficulty remembering what the app can do. Some of these have deep links which means you can launch them directly with the long press on the Windows button.
Battery Monitor w/ Voice Control: (MS – Deep link)
- What is it?:Live Tile for Battery Status.
- Quality: Poor. VoiceCommands hard to find. Do I need this? The app should be faster to return status.
Hey DJ: (MS – Deep link) YYTopic Pick!
- What is it? Music Control, Album, Artist, Genre, Track
- Quality: Great!. App requires that you say the complete names of the artist, song, or album.
- Worth noting: The only app I actually continue to use, written by a Microsoftie.
- Deep link example: "Hey DJ play Born to Run"
Tivo Command: (MS)
- What is it? Speech controlled TiVo remote
- Quality: Good. C&C, requires TiVo account
- Worth noting: Written by a Microsoftie.
ReadyClick : (MS - Deeplink) Y Topic Pick!
- What is it? VoiceControlled Camera Shutter (replaces timer)
- Quality: Fun.Simple app with a high cool factor. Works well.
- Deep link example: "ReadyClick Listen"
Toggle: (MS – Deep link)
- What is it? Control settings for your phone. WiFi, Blue-tooth, Cellular etc.
- Quality: Works. But I'd like be to set the mode rather just go to the control. (ie. "Toggle Airplane Mode Off"). This is a restriction of Windows Phone.
- Deep link example: "Toggle Airplane Mode"
Dictation: These apps focus on voice to text input for the phone. They all have verbal punctuation. Some have connections to mail, messaging and status update. Sometimes the voice service provider is not identified. Biggest weakness is lack of integration with text input scenarios. None have correction interfaces but that doesn't seem to be a big miss.
Dictation Station: (Nuance)
- What is it?: Speech to Text Service for your phone.
- Quality: Fast and Accurate with punctuation! Connects to Email & SMS.
Say Mail!:(Nuance)
- What is it?: Voice dictation for e-mail.
- Quality: Good SR. Can't speak contact name. Doesn't always capture whole sentence. UI not great.
Voice Assistant: (?)
- What is it? Speech to Text Service for your phone.
- Quality: Fast and Accurate with punctuation! English only
Games: These apps focus on entertaining the user.
KungFuStonie: (Could be home brew)
- What is it? Brick breaking game, voice or tap controlled.
- Quality: Both the tap and voice version seem to have latency issues.
Memoraniac!: (MS)
- What is it? Game that tests your ability to recall a digit sequence by voice.
- Quality: Works well. Not an easy game. Could be more fun.
General: Grab bag category
Concept Mapper: (?)
- What is it? On the go concept map creation.
- Quality: Limited speech. Bad UI. Slow.
GPS Voice Navigation: (?)
- What is it? Turn by turn navigation with Voice output.
- Quality: Good. Generic voice directions, ie. No street names. Configurable for Bing or Google as map providers.
Voice Chat: (Nuance)
- What is it? Chat bot with Voice I/O
- Quality: SR is good. UI is poor. Voice output is mediocre. Conversation is insipid.
Voice Toddler Cards: (?)
- What is it?: Flash cards for kids with Voice Output
- Quality: Not very interesting. Likely uses prerecorded speech. Only English and Spanish.
Search: These are specialized search or Q/A applications.
Ask Ziggy: (Nuance)
- What is it? Voice Q&A like. Siri. Does some access to your calendar. Try's to use context to answer some questions.
- Quality: Good, but the latency is not as good as other Nuance based services.
Bingo Voice: (?)
- What is it? Voice assistant.
- Quality: Didn't work when I tried it.
Google(Google)
- What is it? Google search for Windows Phone.
- Quality: It’s g-search. The speech recognition is fine but the UI needs work. Weak feed on when it’s listening. Tends to listen too long.
- Worth noting: It’s the only WinPhone app that features a Google speech back-end.
Urbanspoon: (MS – Deeplink) Y Topic Pick!
- What is it?: Dining Guide.
- Quality: Pretty good. "Urbanspoon find Italian!". Not easy to know what it you can say. Forgot my settings when the app was updated :-(
- Worth noting: This app demos well because the content in the back-end is useful.
- Deep link example: "Urbanspoon find Italian"
Voice Answer: (Nuance)
- What is it?: Q&A system
- Quality: Limited answers. Awkward UI.
Translation: These apps help you communicate in another language.
iSiri: (?)
- What is it?: Translation system with voice output.
- Quality: Seems ok. Not an exciting. Why did they reuse the Siri name?
iSpeak: (Nuance)
- What is it?: Speech to Translation in 15 languages
- Quality: Speech results are slow to come in my test.
Speechy: (?)
- What is it?: Speech to text w/ translation to English.17 languages.
- Quality: Can't contact service.
Translator: (MS - Speech Server)
- What is it?: Speech to Speech Translation (+OCR)
- Quality: Generally works. SR is a little slow.
Voice Translator: (?)
- What is it?: Speech to Text or Speech translator. Many languages.
- Quality: Interface is a little tricky. Speech recognition seems 'good'.