If you’ve read through my blogs you know that I really like my Hello Moshi Clock, and my BlueAnt V1 headset. These aren’t the only speech products I like, and some of the others don’t even use Sensory’s technology. I’ll mention a few others of my favorites, in no particular order:
- Midomi by Melodis. I’ve been playing with this free download on my iPhone. It lets you sing, hum, type, speak, or even hold your phone up to any song and it identifies it. I’m really amazed how well it works when I sing…and my singing isn’t very on pitch! You get to hear samples of different songs that their recognizer calls up. They record some users (it’s a little weird when you hear someone without any musical backing).
- 1800Call411 by Microsoft. This is my favorite of the free information services now available! You can get weather reports, stock market updates, call business, get movie info or buy tickets, find cheap gas, and lots more. The cool thing is that ITS NOT MULTI-MODAL. Multimodal interfaces are great when you can sit staring and playing with your cellphone, but HORRIBLE when driving. Goog-411 is nice too, and possibly more accurate than Microsoft, but business listings alone does not suffice. Microsoft has the right idea…Google is being too conservative (are those crazy young guys getting gray hairs?).
- Vlingo. Another free download for my iPhone. This is the most accurate “say anything” interface I have ever seen. Unconstrained speech is a real bitch, and Vlingo has the first usable unconstrained interface I have ever tried. It usually works. Even when its not 100% accurate it does speed entry. I wish they had a version that was 100% voice interface. I like the mono-modal interfaces where I talk and it talks back, but these are definitely the toughest to conquer.
- Radar by FisherPrice. Maybe I’m being a little nostalgic. This was one of Sensory’s first products released back around 1997, and it may be the best thought out educational play experience of any product we’ve ever done. It was a little robot with a phone. You would talk to it through the phone and play educational games. I can still remember my then 3 year old son, mimicking Radar would say, “What did you say? I can’t hear you. Please talk louder.” It had a true Voice User Interface (that is, there wasn’t a bunch of button and knob backup choices).
- Lightswitches by Voice Signal and VOS. These companies used to duke it out over this market that never fully emerged. VOS used Sensory IC’s (actually, so did Voice Signal in their very first implementation). Later, Voice Signal had its own technology that worked surprisingly well. The challenge in these products was to create a hands free trigger command that didn’t false trigger but worked well in noise. They both kept improving but never quite got there for mainstream acceptance.
I could keep going, but I’ll diverge here with thoughts coming from the hands-free trigger lightswitches. These were a huge opportunity that never seemed to emerge. The products were amazingly handy, and worked surprisingly well, given the technical challenge (especially given the huge success of things like the Clapper, which turns into a strobe light when music is played with a strong snare beat). It was the occasional trigger rejections and false fires that kept these from becoming mainstream. Sensory has been working on this hands-free trigger challenge for about 12 years, and we have a new patent pending approach to dealing with it that will enable a whole new class of hands-free trigger devices ranging from lightswitches to clocks, to carkits, to ovens & microwaves, to picture frames, and other internet connected devices. We should have some interesting announcements with this technology in the months ahead!