are you doing the assessment? If so any preliminary thoughts about the array?

jgan,

I should be receiving the LE1600 sometime early next week. A full evaluation will be posted upon completion. I’ve worked with the M1400, from what I was told, the LE1600 offers several improvements, both internal & external.

Sincerely,

Josh

P.S. Information on our website about the LE1600 can be viewed here:
http://www.voicerecognition.com/motioncomputing/
and the tablet can be purchased here:
http://shop.voicerecognition.com/itemdesc.asp?ic=MC-LE1600

ANY HINTS?

jgan,

Click on the link below to view a article published in a recent copy of PC Magazine.

Motion LE1600 Tablet PC

If you have any other questions, you know where to find me.

Thanks,

Josh

sorry, I had hoped there was info on how dragon and the array are working together. john

John,

We saw a prototype of the new Motion 1600 at a recent tradeshow in Salt Lake City. It seemed like the on-board array microphone did work -or better said- was working well with DNS. I only got a glimpse of the new machine, as the rep was pretty guarded with it, but the new tablet seemed to take the ambient noise of the show in stride, and DNS didn't seem to be freezing up/behaving badly...

The article Josh posted is one of the only articles, and there still is not a lot of public information - that I'm aware of - on this unit...

We'll post more as we find out.

Ingram

Josh, are you close to posting your review of the LE1600?... some of us are really
eager to hear your impressions.

Thanks,

-Coop

Coop,

We are waiting for our unit from Motion...

Ingram

Coop,

I assure you that I will evaluate, then post our findings immediately upon arrival.

Thanks,

Josh

Scansoft has actually created a special spot for the LE 1600 on their site. Under ScanSoft certification it shows the LE 600 as being a tablet with an array microphone that is certified. However using the array microphone they only give two dragons out of five. They say with an externall microphone they give five dragons. I also am very anxious to find how good or bad it works as an array microphone. John

Motion Computing LE1600 Tablet PC

The LE 1600 was tested with Dragon NaturallySpeaking v8 Medical. I tested this tablet using the built-in array microphone, then with a Plantronics CS-50 Wireless USB Microphone. The results are as follows:

Motion Computing LE 1600 using the Built-In Array Microphone set to Narrow Angle
Distance from screen: 14-16 inches

Speech to Noise Ratio: 24





I dictated 1674 words into Microsoft Word 2002, there were 186 mistakes found, this calculates to 88.9% accuracy on the first round of dictation.

Motion Computing LE 1600 using the Plantronics CS-50 USB Wireless Headset
Distance from screen: 16-18 feet

NOTE: I am only evaluating the results of the Plantronics headset with the Motion Tablet to show how the built-in array works in comparison to a USB headset.

Speech to Noise Ratio: 29





I dictated 1677 words into Microsoft Word 2003, there were 19 mistakes found, this calculates to 98.9% accuracy on the first round of dictation.

Specifications of Tablet tested

Manufacturer: Motion Computing
Model: LE 1600
Operating System: Windows XP Tablet PC Edition
Processor: Intel ® Centrino ™ Mobile Technology with the Intel® Pentium ® M Processor Low Voltage 758 (2MB of L2 cache, 1.50GHz, 400MHz FSB)
Chipset: Intel 915GMS
Display: 12.1” XGA TFT LCD with viewing angles up to 180 degrees
System Memory: 512MB 400MHz DDR2 memory (Upgradeable to 1.5GB)
System Storage: 1.8” Hard Disk Drive (HDD) with 30GB or 60GB capacities
Audio: Motion Speak Anywhere™ technology. This multi-directional array microphone design intelligently switches between two of the three microphones based on screen orientation. It uses Knowles Acoustics IntelliSonic software.
Security: Integrated Fingerprint Reader. Universal lock slot and Motion Security Center
Weight: 3.13 lbs.
Battery: Lithium-ion battery with 35WHr capacity

Thanks,

Josh

Thanks Josh,

Well, the array microphone performance is disappointing... but, not surprising. Did you test with respect to screen orientation?... any difference?

What is your overall impression of the LE1600 as a tablet? Does CPU speed seem adequate for DNS? Any conflict with power saving features?

Does the Plantronics CS-50 receiver require line power or will it operate from the USB connection or battery operated? Looks like an ideal companion for tablet operation provided you don't have to power from the mains.

Finally, what is your technique for including screen shots as attachments?

Thanks in advance,

-Coop

PS Any other anecdotal impressions or specs are always appreciated.

Thanks Josh!

Wish I had a copy of your script (Words) for attempted reproduction?

Coop,

The screen orientation made little difference. I wish to include that when I tested the array element, I was in a typical work environment, people talking in the background, phones ringing, etc… I'm certain that in a quiet environment, the array would have produced much better accuracy.

Overall, I thought the LE 1600 was a great tablet. I had no problem with installation, training, or dictating. The CPU speed was adequate. When evaluating a product. I make sure to test as it would be used in “real life”. I dictated into Microsoft Word, I had Outlook, Excel, & Internet Explorer running in the background. I did switch between several applications, the tablet kept up wonderfully.

The Plantronics CS-50 USB wireless headset is amazing, however; it does require a power outlet. The earpiece is extremely light; I often forgot that I was wearing it. I also found that I could stand approximately 60 feet from the unit and achieved flawless accuracy. I highly recommend this headset for any PC need.

As far as screen shots are concerned; images can only be added if you have a URL for that image. As you know in order for an image to have a URL, it must be ‘web-enabled’. We simply created a site where we could post our images.

Thanks,

Josh

I read excerpts from, The Calcium Bomb by Douglas Mulhall & Katja Hansen. I was able to evaluate the Motion LE 1600 while learning about the nanobacteria link to heart disease & cancer. Very interested book!

Thanks,

Josh

FYI:

Based on our reviews and evaluations of this new tablet, we have decided to carry this item. This will be the only tablet pc that we offer.

The Motion LE 1600 is a great tablet for use Dragon NaturallySpeaking.

We will have these tablets available shortly.

If you have any additional questions, please do not hesitate to call.

Thanks,

Josh

Josh, would you mind testing the tablet in a quiet area? I would very much appreciate how this works with the array without external noise influencing. tia, john

Interesting. I use a Toshiba M200 with the 3 microphone array system. In fairly quiet environments, I have very good success with 98% accuracy using the array microphones, IF I use the computer in laptop mode. The accuracy drops off when I use it "turned around" in Tablet mode, especially if in portrait position. I'm not sure why.

This is a bit inconvenient, as when I'm in a patient's room, I use it in Tablet mode, then in between patients need to swivel the screen to dictate, but nonetheless, it works great and I go without the need for a wired headset of any kind. I work very efficiently this way.

Accuracy drops off in noisy environments. It does NOT need to be silent, but it is helpful for me to dictate in my own office, with the door closed.

I was about to purchase the LE1600 for an incoming physician, but now I wonder...it would be nice to compare apples to apples and know about accuracy of the 2 units, in the same conditions. It also would be nice to know why one is better than the other. Does it have to do with the arrays themselves being different, or perhaps the noise cancelling software? Toshiba has "Toshiba Audio Effect" software which does something in the background, and it is very important to have this running to have the array system work well.

Joe Graney, MD

You ask, and you shall recieve...

This time I tested the Motion LE1600 tablet in a quieter environment. I set the built-in array microphone to Medium angle, then I Calibrated the Mic. ( A simply process done within the Motion Dashboard).

I created a brand-new profile (without anaylizing my writing style). I dictated using Dragon NaturallySpeaking Medical v8. My results we as follows:

I dictated 1224 words into Microsoft Word. There were 48 mistakes, which calculates to 96.4% accuracy on the first round of dictation.

Speech to Noise Ratio: 25





This time I attached my dictation. I had Microsoft Word track the changes to show what type of misrecognitions Dragon NaturallySpeaking made. The Image below will show two things. The word underlined in red or blue is the intended word. The word on the right, after 'Deleted:' was the misrecognition.

Example:



If you have any questions, you know where to find me.

Thanks Josh,

I (we) appreciate the additional testing. Looks like the array mic is at least usable in quiet envirionments.

-Coop

Josh, I looked at the attachment and found a number of additional errors. I know that you put in quite some time on this (which I appreciate.) With the additional errors counted in I think the accuracy would drop significantly. Probably at least several points. To me at that level of accuracy, it is questionable whether it could be used with an emr on a regular basis. I think that others have said at 95% or above you are ok, anything less and frustrations overtake. john

John,

I will gladly re-dictate using the Motion LE1600. You’re right in that there were additional errors (not corrected) in this document. The reason: Upon ‘Play that Back’ I realized those were errors on my part (fumbled speech, etc.), not Dragon.

My second dictation into the tablet using the array microphone was even better.

As you know Dragon improves with every use. In my second dictation, my accuracy improved by .4 points. This time I dictated 1061 words, there were 32 mistakes, which calculates to 97.0% accuracy.

I’ve attached my dictation so you can see exactly what kind of mistakes were made.

If you have any additional questions, please do not hesitate to ask.

Thanks,

Josh

Josh,

I took the opportunity to review your more recent document on the LE1600 accuracy, and found an occasional homonym that was mistyped. For instance, in a couple of situations you said "person's" and it was recognized as "persons". To be honest, the majority of my medical transcriptionists that I'd used over the years made mistakes such as this on a regular basis, so I can understand how this can be overlooked.

Analyzing the accuracy, including mistakes such as this, still offered around a 96% accuracy on the 'Motion LE 1600 DNS Eval' file. Understanding that the average transcriptionist is alleged to offer no more than a 98% accuracy, this is pretty impressive, in my opinion.

I think that a number of forum visitors would be interested in hearing the results of similar dictations in landscape vs. portrait mode, with the Motion Computing LE1600 Tablet PC, using the array microphone.

If, in fact, you could do this with a medically oriented dictation, that would be even better. If you need a sample document to read, let me know.

Thanks

Eric 21,

I will be onsite today at two different medical facilities, I will be testing how this tablet performs in a 'real-life' medical environments.

I will evaluate both the portrait & landscape modes.

This is an amazing tablet, I’m confident that it will meet everyone’s expectations, then some.

Thanks,

Josh

Josh,

Thanks for your evaluation of the LE 1600 and Dragon. Kudos to Motion for a reasonably accurate mic array system. Besides the accuracy I am also curious about the speed of DNS 8 on the LE1600.

How fast did you read your script for dictation? Was there any lag with Dragon procesing your dictation into words? How does speed compare to other computers ? I have tried the Toshiba M200 2.0 Ghz with pretty good results in both accuracy and speed with DNS 8 Pro. Also, how quickly does Dragon deploy voice commands in the LE1600 ?

Thanks.

Ron,

I was very impressed with the speed & accuracy of this tablet. I dictated at my normal pace (which is pretty fast), it transcribed fairly quickly.

The tablet I'm testing only has 512MB of RAM, therefore; I noticed a short delay. I am certain that if I had 1 Gig of RAM, the transcription process would have been instant.

Thanks,

Josh