I’m getting more and more impressed with the Fujitsu P1610 each day. Tonight I took the opportunity to play around with the speech recognition using just the one external microphone on the device. Actually, I guess it’s an internal microphone that I’m talking about. Fujitsu has incorporated some advanced array microphone technology into the P1610, even though it only has one microphone. This technology is pretty cool and seems to enable a single microphone to perform much more like a dual array microphone. I decided to give speech recognition using the Tablet PC recognition engine a try tonight and so far I am pleased with the results.
The first thing that I did was to complete the basic voice training exercise that comes standard with the Tablet PC. I would normally recommend that a user complete all of the training exercises up front to get the best accuracy possible but because I was short on time I decided to try using just the one training session. I must admit that I wasn’t expecting too much on the accuracy front, but the result has been quite surprising to me as it’s been a lot more accurate than I thought it would be.
The first thing I did was to tune the microphone using the included IntelliSonic software application. This program was included by Fujitsu to take best advantage of the single integrated microphone on the P1610. I set the array to be a narrow focus area directly in front of the microphone to better enable the speech recognition to work. This entire post was generated using the speech recognition and it only took about five minutes to create in total. I had to do a fair bit of correction but only very simple corrections as a few of the more specialized words were not recognized correctly. This was not surprising to me because words like P1610 are not normal words in the English language. To give you an idea of the accuracy the following paragraph will be posted unedited, just like it was recognized by the computer.
In so far I’m really enjoying using the Fujitsu P1610 Tablet PC. This small slate form is working out very well for me on a daily basis, and I find it provides the maximum utility with the minimal amount of BULK and WEIGHT to carry on a daily basis. the words appearing in all caps are words that I had to spell for the recognizer to get correct. It is really nice that you can spell words in the middle of sentences to have them peer correctly.
I am looking forward to trying Dragon NaturallySpeaking 9 using this integrated microphone. I believe it will do a very good job with much less training. Next up I am going to try the system command control features and I’ll report what I find using those in a future post.