GDC 2013 recap
This blog is brought to you from Kevin Griffin, the AT&T Developer Program Technical Biz Dev/Product Manager, who just got back from GDC:
Some partners in crime at the AT&T booth include (from left to right): Giri Bhaskara, Glen German, and Patrick Ehlen.
My passion and background in game development runs pretty deep as I previously worked several years at Microsoft on their DirectX API Platform which enabled games to have efficient fast access to the hardware that accelerated graphics and sound on PCs. DirectX was the precursor to the original XBOX program (the X was used as the basis of the name Xbox to indicate that the console was based on DirectX technology). Several of the same people I used to see at the original DirectX Meltdown events have maintained their involvement in games all these years and it was great to see them again at GDC. Many late night LAN gaming sessions were spent “testing” the original DirectPlay back in the day. If you’d like to read about the history of games in Windows, a great starting point for those interested in the foundations of Windows gaming/DirectX is a wiki article about DirectX and its origin here http://en.wikipedia.org/wiki/DirectX .
Microsoft was also producing a simulation title at the time called, Microsoft Flight Simulator, affectionately nicknamed “MSFS”. My first days spent working on the Flight Simulator and Combat Flight Simulator franchises were just as the team was getting close to shipping Microsoft Flight Simulator 2000 in 1999. MSFS 2000 was the last version of Flight Simulator that virtual pilots flew around in a silent world. As the Aircraft and ATC program manager I designed an Air Traffic Control system that used real time text to Speech concatenation with ten different voices to produce a living world. The system had the performance scalability so that if you tuned your radio in the aircraft, you’d not only hear other aircraft taxing around on the ground and flying in the airspace around you, each aircraft would be using a different voice. The MSFS speech engine made for a very powerful immersive simulation of real life flying and was first launched as a part of Microsoft Flight Simulator 2002. At GDC I had previously presented for DirectX, and the Flight Simulator Speech SDK and “how to” for developers to customize the simulation with their own voices prior to AT&T’s presence at GDC 2013. One of the questions at our Q&A at our GDC Session on “The AT&T Speech API: Bringing Advanced Speech Technology into Play” was about speech concatenation, and resulted in flashbacks where I spent hours upon hours recording and re-recording various words in the grammar that went into just one voice pack (used to concatenate speech) in Microsoft’s Flight Simulator.
One of the founders of AT&T, Alexander Graham Bell, or as his close relatives and friends called him, “Aleck”, had a passion for simulated speech. Long before the telephone was invented and patented in 1876, Bell’s father took him to see a unique automation, the “mechanical man”, which simulated human voice. This was in 1863, and this fascination with artificial speech sparked the interest which eventually produced the telephone, the founding of AT&T and creation of several other speech related inventions. This common core interest in speech technology has remained a passionate genetic thread for AT&T researchers ever since.
Fast forward to GDC 2013. A few weeks ago I learned the timing of our release of a new Gaming context in the Speech API, was lining up with GDC 2013. I was compelled to present a business case for AT&T’s presence at GDC, and management agreed. Well, as you know, you aren’t anything at GDC unless you have something compelling to show. So the challenge was on…
After presenting the idea of combining a few different areas we’d been working on, that developers would be interested in, we ended up was mashing up our Microsoft SDK with an existing Unity3D Sample, to demonstrate Speech to text.
Although we just had a few days to learn Unity3D, and pull a compelling demonstration, I think you’ll agree after downloading the sample here that it was executed pretty well. A couple photos of the sample running are below. You can, completely hands free, just by talking to the sample; change the characters, change their hair, shirt, shoes and also tell them you like their shoes or shirt and they will animate their looks completely. The sample is entirely powered by the Speech API.
Focusing on the Unity sample which you can see running in the background in the photos above drew a constant crowd to the booth.
It’s no doubt interest in Unity’s game engine has grown dramatically in the last few years, just looking at their booth size alone, you can tell they’re doing very well.
Prior to the show, I hadn’t used Twitter at all. So I was given access to the AT&T Developer Program’s twitter account at GDC to help with our online social presence covering the event. My first tweet ever was retweeted by the CEO of Unity, to which I now receive a constant hard time from the team.
In addition to AT&T’s presence in on the expo floor, Patrick Ehlen and I also presented a session titled, “The AT&T Speech API: Bringing Advanced Speech Technology into Play”, as a part of the Smartphone & Tablet Games Summit track. The session went very well, was well attended and we had an extensive Q&A talk. At the end of the session we drew business cards of the various people in attendance and the winners were:
Francois Desmarais – LG Optimus G
Raymond Fontenot Jr. – LG Optimus G
Nick Laing – Prepaid AT&T card
In addition to downloading the sample Unity 3D + Speech demo above that was shown in the session and in our booth, you can contact me for a copy of the PowerPoint we used. It truly was a pleasure being at GDC again, and representing a pretty cool product.
Thanks for reading my blog, and feel free to ping me with any questions or comments below or via twitter @H2DRIV3R.