Reaction from the Backstage event addressing speech recognition in WP7


Saying that the just concluded event was a let down would be an understatement of the highest degree!  For an interview that they had planned in advance, they came out ill prepared and had to go through several phones before they found a working one to demo the features! Microsoft’s Liane Aihara, gave a disjointed presentation that offered nothing new or just a possible timeline of when some desired voice command features would be available. In fact, the opposite was true because she clarified more features that would NOT be available at the launch. I think that even she knew she had done a bad job because at the end of the surprisingly short interview, she quipped “that was terrible!”

The basic premise for speech recognition on WP7 as it now stands, will focus on three areas where consumers would most likely find convenient namely: Call a contact, launch an application and search (via Bing) There were moments when I was screaming at the screen shouting, why not just port Voice command to the OS? Who did this focus group about the  commonly used voice features because they should not have job! So here is what will be possible with speech recognition 1.0 in WP7.

  1. Call a contact
  2. Launch an application
  3. Do a general search via Bing.

That is it! The search functionality will be handled via the cloud while the “Call contact” and “Launch application” will be handled locally. The only new thing  I learned was that a user could use speech recognition even when the phone was locked by enabling it in the settings. It will be available in English (US/UK), German, Italian, French and Spanish for the first two features and only in English for search. She was proud to say that there would be no need to train the app for voice recognition for non native English speakers and that they had received good feedback from that group. Since the app is native to the OS, it will be upgraded via firmware updates and not as a separate app.

And now for the downer part. I suggest that you if you are having a bad day, you may not want to read this.

Things that you can’t do.

  • You cannot search locally on the device
  • You can use it to launch the Zune app but then you have to manually control it after that. 
  • You cannot  make it play a play list.
  • Will work with Bluetooth but only with the 3 features listed above
  • Will not have any translation capabilities
  • Voice navigation was never mentioned
  • No correction capability once the command has been uttered. You have to repeat the command.
  • No ability for 3rd parties to select data from a list box or to fill text boxes with speech i.e  speech to text. (This is also not available for native apps as far as I know)

I am not sure why they even had the event and considering Google’s announcement of the new Voice Actions for the Android, I expected so much more. Not to say that I will not be getting a WP7 phone but this was definitely depressing. Come on Microsoft, you can do better than that! You already had a great Voice command  feature in WinMo 6x and  great voice note taking using Microsoft Recite. Even windows 7 OS has better voice capabilities! What I would like to see is for Microsoft to give a general timeline of what features will be added and approximate times for their arrival.