I believe it was to be today, once upon a time...

It certainly WAS (once upon a time)…

We are now estimating the delivery to be towards the end of October because of some last-minute design implementations that generally everyone felt should to be added

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Would anyone be willing to bet hard money that it actually will ship by the end of October?  The ship date has been receding since what, March or April?

Nope.

Lunis, if design changes are still occurring, how can a ship date be estimated? Don't design changes have to implemented, tested, and then manufacturing can start only afterward. If the tests fail, then more redesigning is needed, etc.. 

The Speechware website hasn't changed their ship date from mid-October.  Strange.  http://www.speechware.be/en/verArticulo.php?id=1627  Lunis, are you able to discuss what these design implementations are.  I'm kind of interested.  Thanks.

If the SpeechWare 3-in-1 microphone were to be shipped in its current condition, it would still best every other TableMike on our website but the beta testers and SpeechWare would like to do better than that. Simply put, we know the microphone isn't as good as it can be and if delaying the shipping date by 2 weeks would make a difference, we were all in favor of it. The devil is in the details.

SpeechWare simply hasn't updated their mid-October release date on their website.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Please, introduce us to this horny fellow and tell us those details. 

Matthew et al.,

Basically, it's a long story for which going into the gory details won't make for anything but a long post. However, I'll try to keep it as simple as possible.

First, any time that you develop a new technology there are many unanticipated consequences and no amount of testing can catch them all. In this particular case, the problem was that the SpeechWare 3-in-1 uses what is called Automatic Gain Control. What this does basically is to:

1. Set the amplitude (voltage, gain, volume – whatever you want to call it it's all the same). This is dependent upon your distance from the microphone.

2. The secondary function of the AGC is to basically turn the microphone on when it detects an output signal of sufficient amplitude such that Dragon starts analyzing the output signal and then turn the microphone off when the amplitude (volume, gain, voltage) falls below a certain threshold. The purpose is to prevent feeding output continuously to Dragon. In essence, the purpose is to confine the recognition process to speech only.

This is a tricky process. Back in the spring and early summer the base units submitted for testing purposes had a problem with the AGC. The problem was basically that the microphone wasn't turning on quickly enough, with the consequent result that there were dropouts when beginning to dictate. What this essentially did was miss words or confuse words with the consequence that unless you were dictating continuously accuracy went to hell in a handbasket. The problem was then submitted to the SpeechWare engineers, who subsequently found the problem and fixed it. A second prototype base unit was then submitted for testing. Preliminary testing found that the problem with dropouts was fixed, so the base unit was finalized at that point and a complete product prototype was submitted for testing in the beginning of September. Unfortunately, the test samples got hung up in customs and SpeechWare had to make a decision (i.e., go or no go) with regard to production. When the final production prototype was received and available for testing, we found that there was a problem with the microphone sensitivity such that the AGC was turning on properly, but not shutting off in all cases because the cutoff threshold wasn't matching the microphone performance in terms of signal output.

While this wasn't a problem with DNS 10.1, DNS 11-11.5 changed the rules of the game with regard to how much data could be stored in the speech buffer and processed in parallel (multithreading) using two Acoustic Models. The consequent result was that because of the distance from the microphone, the sensitivity was simply too high and the output signal was picking up background and feeding it to Dragon 11-11.5 on a more or less continuous basis. While I laud Nuance for attempting to create a multithreading version of Dragon NaturallySpeaking, they introduced some new features and one major bug in their multithreading code, with which some users are currently experiencing a lot of latency issues that should not be occurring. There are a lot of reasons for this but suffice it to say that the end result for the SpeechWare 3-in-1 microphones and algorithms was that the microphone was tending to remain on instead of turning off because the entire system was too sensitive to background sounds.

Therefore, having discovered this particular issue, the problem was resubmitted to the engineers along with a specific test scenario for their testing. The consequent result was that they: (a) found that the microphones currently scheduled to be shipped with the final product were simply too sensitive to background noise levels, even at a moderate level of background noise and, (b) as suspected, the microphone sensitivity had to be reduced so that acceptable background noise levels (volume) wouldn't prevent the AGC from turning the microphone off when and where appropriate. Obviously you cannot, and will never be able to, dictate using the SpeechWare 3-in-1 while setting in the first row of a rock concert. That is simply the nature of all microphones, including the best noise canceling microphones. Even the Sennheiser MD 431 II, although still virtually 98% accurate, would still cause some delays and reduce the accuracy by at least 1% using Dragon NaturallySpeaking 11-11.5 under that condition. Regardless, the goal was to prevent certain levels of background noise from keeping the microphone turned on and continuing to send output to Dragon when, at least within the design and acceptable specifications possible given that the SpeechWare 3-in-1 is a desktop microphones vs. a close talk microphone, the AGC should turn the microphone off when the user stops speaking (pauses). There are a couple of important points to take into consideration:

1. AGC is not a new technology or new concept. It has been around for many, many years. Its primary implementation is in handheld voice recorders (analog and digital). In that arena it is termed VAC (Voice Activated Control). However, in hand-held recorders, it doesn't have to be as responsive (performance – speed of on/off performance), and the microphone sensitivity does not have to be as finely tuned as it must be for speech recognition. That is what's new in the SpeechWare 3-in-1.

2. Even under the current conditions with the current prototype and microphones, accuracy is NEVER an issue. The accuracy with all the algorithms in place and functioning is still virtually 99% or better, even at moderately high noise levels. Even so, the problem with sufficiently high background noise levels is that performance Re: latency is going to slow down the response time if the background noise levels are sufficient volume (amplitude/gain) to keep the AGC engaged. Users are simply going to have to understand that the SpeechWare 3-in-1 is going to experience latency issues if the user abuses (uses the microphone outside of the design parameters), such as trying to dictate in an airport terminal or crowded train station, or, God forbid, at a rock concert. The SpeechWare 3-in-1 is simply not designed to handle that level of background noise. In addition, door slams, telephones ringing, listening to CNN, listening to music on the radio, or dictating when someone is looking over your shoulder or in the approximate vicinity and talking is going to have an impact on the SpeechWare 3-in-1. However, it's never going to affect accuracy, only performance (latency). The only thing that's going to affect accuracy with the SpeechWare 3-in-1 will be the standard slurring words, running words together, enunciating unclearly, and the normal pattern of user controlled faux pas that are anathema to speech recognition accuracy. The bottom line is that if you use the SpeechWare 3-in-1 within the design parameters established and understood, then the SpeechWare 3-in-1 will perform as designed. But, the ANC (Automatic Noise Control) will continue to prevent the SpeechWare 3-in-1 from interpreting any type of background noise as speech. The bottom line is that accuracy, regardless of how the SpeechWare 3-in-1 is used, will continue to remain as good as or better than any current designed for speech recognition microphone. It just may take Dragon longer to process your speech if you abuse this microphone by dictating in a noise environment for which it is specifically not designed to handle (volume/amplitude/gain).

3.  The current problem, which should be resolved by the new microphone units, is primarily due to problems inherent in DNS 11-11.5, particularly with BestMatch IV, but, though less significant, even with BestMatch III to a certain extent. On the other hand, the problem does not exist with DNS 10.1 even with the current microphones that we have for testing.

4.  SpeechWare and the SpeechWare engineers will continue to improve the SpeechWare 3-in-1 on an ongoing basis just as is done with any hardware/software technology. Also, SpeechWare has a lot of irons in the fire about which it would be inappropriate to speak at this time.

We should be receiving the test microphones shortly. If these adjustments that have been made solve the problem, then this should not prevent the SpeechWare 3-in-1 from shipping at the end of October. Cross your fingers because we still need to test these under real world conditions before shipping the final product.

Regardless, the entire process of producing a desktop microphone of the quality that users will find in the SpeechWare 3-in-1 TableMike™ has been a long and difficult path. Having been an integral part of this process, it behooves everyone to continue to be patient. Software/hardware development and implementing new technologies is not something that is either easy and/or without its own unique set of problems. I think everybody understands because we can all see the unfortunate consequences surrounding the release of DNS 11-11.5 and the amount of time that was required before Nuance released the 11.5 update. The entire process has to be carefully done and carefully tested. Please do not assume that any lack of careful or or precise design/development/for testing was at the root of the any of the delays in getting the final product out the door. Nothing could be further from the case for the truth. SpeechWare has spent more time, money, and care to ensure that the SpeechWare 3-in-1 lives up to its claims in terms of performance and accuracy. Everyone involved understands the frustration that everyone who is chomping at the bit to get their hands on this particular microphone has undergone as a result of this process. Nevertheless, everyone has to understand that the whole process of development of any type of new technology is not something that you can simply pour into a mold and pop it out for sale and distribution any time that you want. It's complex, it's difficult, it's involved, and not everything works perfectly or falls into place exactly as everyone would like. I'm sure that the SpeechWare folks would be the first to acknowledge and understand everyone's excited anticipation and frustration over the delays. So, please don't be too hard on these folks. Barring hell freezing over and the creeks rising, which we don't currently anticipate, everyone should be happy campers shortly.

On behalf of SpeechWare,

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make speech recognition, we make it better!

"Good enough never is." - Debbi Fields: founder of Mrs. Fields Bakeries

-------------------------

It seems this product is as much (or more) firmware as it is hardware.  As one of the people with a pre-order in, and in light of this most recent delay, can you remind me if the firmware is user-upgradeable?  The USB connection should allow a pretty easy upgrade if the microphone has been designed to allow this.

 Thanks!

-------------------------

DNS 11.5 Pro / VoicePower
Work:  3-in-1 TableMike USB / Win7 64bit / i3 550 / 8GB
Home:  Sennheiser ME65/K6 / Win7 64bit

This is one of the irons in the fire so to speak. I recommended this to SpeechWare a long time ago and they generally agree. However, this has to remain on the back burner for the time being. In any event, it would be premature to talk about what can or can't be done currently with regard to a firmware upgrade. It would be inappropriate to discuss anything without a set of pure technical specifications from the engineers that clearly identify what they can do and how they can do it. Is it possible, absolutely. Is it possible with the current system boards, that is an unknown at this point and there really isn't anything that anyone can shed light on.

It would be nice to say yes, no, maybe, etc., but it's a long way from possibility to implementation. Simply put, there's nothing to talk about at this point.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make speech recognition, we make it better!

"If the automobile had followed the same development cycle as the computer, a Rolls-Royce would today cost $100, get a million miles per gallon, and explode once a year, killing everyone inside." -- Robert X. Cringely

-------------------------

I am very impressed with Speechware and there concern to iron out issues prior to distribution.  Given this follow up I would rather wait a month etc if necessary for a product that works correctly.  I am impressed they are willing to fix an issue DNS 11 created.

 I only hope being an early adopter wont lead to being left out if there is a future firmware upgradable microphone.  the fact this is possible makes wonder if I should cancel the pre-order and wait 6 mo to see if this is available.  I hate getting a new product to see it revised shortly after production and being left without a key feature.

 As usual Nuance, has created an issue that I am sure they know exists and have done nothing to fix it with an update. I can only hope someone else will enter the speech recognition arena and provide actual customer service and support for the product.

I wish I knew why I buy the new versions, there are bugs they create that are never addressed.  they simply are one of the worst companies for support.

 SpeechWare, has impressed me and should be applauded for what they are doing.

Chuck, as a customer, let them know, I for one really appreciate their efforts to provide a bug free product and to care about the end-user.

Neil,

 I totally agree with you about Speechware holding off shipments for improvements. I used to resell mid to high end audio electronics and had some manufactors that did this when they saw a problem. Most "other" manufactors would just release their products to gain sales. Usually higher end products do care about quality first and quanity second.

Nice to know this exists in the speech world,

Larry

Likewise, I am impressed with SpeechWare's consideration of the customers need and willingness to address problems before release of the new 3-in-1 mike.  But I am also concerned about whether to cancel my order and wait for the "final" version - updateable with firmware. For the time being, I am quite satisfied with my SpeechWare 2-in-1 mike until the new one is fully tested and ready for market.

What say you all? 

-------------------------

DNS 12.5 Pro, Win7, i7-3770K @ 3.7GHz, 16 Gb, 256 Gb SSD Boot drive, 3-in-1 TableMike - KnowBrainer 12 - Thanks so much, Lunis.

 

 

 

 

 

 

 

Part of the reason for the additional wait time is so that the 3-in-1 really will be finished when it arrives but there is certainly no problem with refunds as we only have 8 remaining units for pre-orders to account for.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Thank you, Lunis, for the comment.  With your assurance that the 3-in-1 will really be finished when it is finally shipped, I'll stay on the pre-paid list and wait to get the finest desktop mike for speech recognition.

 So what is the latest timeframe from SpeechWare for deliveries? Any word?

-------------------------

DNS 12.5 Pro, Win7, i7-3770K @ 3.7GHz, 16 Gb, 256 Gb SSD Boot drive, 3-in-1 TableMike - KnowBrainer 12 - Thanks so much, Lunis.

 

 

 

 

 

 

 

It isn't solid but it looks we will be able to ship out between November 7 and November 14. WpeechWare is in the process of making 1 additional tweak before shipping. As of now, we have presold 100 out of our 108 incoming stock.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

I want my 3-in-1!

Between November 7 and November 14 would qualify as "towards the end of October" under a broad view.  But that is probably as flexible as the previous dates.

Lunis:

Any update? I'm looking forward to getting the mic soon.

-------------------------

Dean L. Surkin

DNS 11.5, Windows 7 64-bit, Intel Q9500, 6 GB RAM

Welcome to the KnowBrainer Forums

It now appears that will be shipping between November 14 and the 21st. SpeechWare is applying 1 final tuning to the microphone.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

At least we narrowed it down to what month........

 Tune away baby. I want it to be stellar!!!

Thanks,

Larry 

Of what year?

G.J.,

2045.

Just kidding.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make speech recognition, we make it better!

Be careful what you wish for.  You may find that what you get is not what you expect, or what you want. - Aesop (620 BC - 700 BC modern interpretation)

-------------------------

Maybe we'll be lucky enough to get the 3-in-1 mike in time for us to be Thanksgiving for what we have received.  If not, we'll be thankful for what we have - life, liberty, love - Chucker and Lunis - and DNS. .... more and more.

-------------------------

DNS 12.5 Pro, Win7, i7-3770K @ 3.7GHz, 16 Gb, 256 Gb SSD Boot drive, 3-in-1 TableMike - KnowBrainer 12 - Thanks so much, Lunis.

 

 

 

 

 

 

 

Many thanks, Chuck. I'm glad to know that the test-engineering-test cycle will complete.

On a related note: Do you know of anyway to keep DNS11.5 from having the latency issue (due to multithreading), i.e., a work around with  our current mics? Does changing the input volume to some optimum value work, or is this a bug that just needs fixing by Nuance?

 

For experimental purposes, we created a new user profile utilizing Best Match III and liked it so well that we made the change permanent. Although we were not experiencing any latency issues, we wanted to try it with the SpeechWare 3-in-1 because of the earlier oversensitivity issue.

Unfortunately you cannot change a Best Match IV profile to Best Match III. You have to create a whole new user but if you're using 1 of the professional versions of DNS, you can export your vocabulary via the Manage Vocabulary menu, export your commands and reverse the process for the new profile. This will not only eliminate the need to train but should give you a nearly identical user profile utilizing Best Match III.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

I found that using a BMIV user on a uniprocessor machine worked--with a message from DNS that it was running in bm3 mode. Did you try that?

 

We hadn't considered that approach but it certainly sounds clever. It's interesting that Nuance anticipated someone copying their multicore profile to a netbook or older processor. Nice catch

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

How is this done?

-------------------------

Dell Inspiron 7520 SE,  2ghz core i7(Ivy bridge), 8gb ram, 750gb Seagate Momentus XT SSD/HD Hybrid hard disk. SpeechWare 3 in 1, Sennhesser ME3, Buddy USB 6G, Windows 7 Ultimate, Office 2007.

I remember a few ways. One is to import an old BestMatch III profile into DNS 11. Another is Lunis's suggestion to create a new user profile and click the advanced button to select BestMatch III, and after the new profile is created, import commands and vocabulary from a good  BestMatch IV user profile. Another way is to boot Windows into uniprocessor mode, though I don't remember how to do that.

Finally, the easiest way is to run Windows on a virtual machine (e.g. VMware workstation) and set the number of virtual processors and cores to one.

 PS, found it. Run MSconfig, click the boot tab, then the advanced options. You'll see a small window that allows the number of processors and maximum memory to be determined for the next boot.

Glad you like it. You've inspired me to compare the best match III and IV. I do get delayed transcription when I'm trying to edit a large document in Microsoft Word 2003 using BestMatch IV, so I'll give III a chance (though I remember several discussions pointing the finger at Microsoft word rather than DNS for this kind of a problem).

The data are in. The long delays between an utterance and the transcription that I see in large documents in Microsoft Word is not affected by using either BestMatch III or IV. In short word documents, or in DragonPad, both speech models are equally quick as far as I can tell. In long word documents, both have a delay that we have attributed to either Windows or Microsoft office in previous threads.

Just happened to notice the Speechware website now says shipping by Dec.  

 http://www.speechware.be/en/verArticulo.php?id=1627   

Lunis,

Still shipping starting on Monday????

Also do we get the cool Monoaural Earphone???

Larry

We believe the ear phone will be included with the final product.

SpeechWare sold out of their original November inventory and will not be able to restock until, most likely, mid-to-late December. HOWEVER, most, if not all, of the beta testers found the 1st generation of the 3-in-1 to be a bit too sensitive. Although the oversensitivity did not produce unwanted dictation it did cause latency delays in DNS 11 because DNS 11 includes a feature that comes into effect when NaturallySpeaking isn't certain if it heard actual dictation or noise. This feature typically takes approximately 4-5 seconds to decipher each random noise so that if you squeak your chair, press a key too hard and purse your lips, NaturallySpeaking could produce a latency delay of up to 15 seconds while it deciphers noise from words. That translates into a 1\4 minute delay before your dictation hits the page. Keep in mind that up until recently, McDonald's was guaranteeing a 15 second delivery time at their To-Go Window! Quite frankly, we would like to see this NaturallySpeaking feature eliminated because it seems to be what we hear the most complaints about. We believe that it would have been better to let the hardware (sound card or microphone) handle more of this work rather than NaturallySpeaking but we can certainly appreciate what the DNS developers intended.

Because of this NaturallySpeaking weakness (for lack of a better term), we requested that SpeechWare reduce the sensitivity of the 3-in-1. We have been experimenting, over the past several weeks, with altering the sensitivity on numerous prototypes. On November 11, our company and the SpeechWare development team independently came to the same conclusion. We believe we have found the perfect balance and have been dictating all of our forum postings and e-mails, since November 10, with the final release product. However, the microphone elements still need to be tweaked, by the Belgian team, early next week. SpeechWare should be able to ship to the US around November 16 and we should be able to get the product out to our customers on about November 21.

Although SpeechWare has not received any complaints on the 1st (non-reduced sensitivity) 40 units, which they sold locally, we felt our customers would rather wait a few extra weeks for us (referring to Chuck Runquist, Larry Allen, Jon Wahrenberger, James Carr, Rüdiger Wilke and others, whose names we should've remembered) to help SpeechWare fine-tune the technology. Although we are currently not permitted to post any of the early customer 3-in-1 feedback, we have read a number of the reviews which were all extremely flattering.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

"the microphone elements still need to be tweaked, by the Belgian team"

 Does this mean that each time, during the many "tweaks" described in the last several months, a team of elves has to go and adjust each  of X hundreds or X thousands of mikes waiting to be shipped?  That would take time.

Reminds me of Midway, where the Japanese had armed the planes with a certain type of bomb, and we ready to go, but before they could take off from the carrier,  decided that they had to arm them with another type of bomb (torpedo vs. regular, forget which order). Had to take planes below deck to rearm, and got caught during the change.

Less than 300 microphones will need tweaking.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

January?

Don't panic Your microphone should ship during the week of Thanksgiving but we have sold out of our original allotment and started taking pre-orders for the 2nd allotment which is due on January 2.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

This is a little troubling. In just a few days, the ship date went from November 21st to the week of Thanksgiving. This means that for the past few weeks, the ship date has slipped by one week each week.


-------------------------

Dean L. Surkin

DNS 11.5, Windows 7 64-bit, Intel Q9500, 6 GB RAM

Dean,

Your frustration is understandable. However, consider the fact that DNS 11.5 was released just two months short of a year from the original release of 11.0 and that was just software (August 12, 2010 vs. June 10, 2011). Hardware is a whole different can of worms. We are all anxiously awaiting the finalized SpeechWare 3-in-1.

What would've happened had we not found the things that needed to be adjusted? Asking for everyone's patience at this point is a bit overcooked. However, SpeechWare is doing everything that it can possibly do to ensure that the SpeechWare 3-in-1 gets here working as it was originally promised. All I can say is that when it finally arrives, you won't be disappointed and you will be thankful that everyone involved, including and especially SpeechWare, has worked overtime to make sure that the SpeechWare 3-in-1 performs up to its claims.

All I can tell you is that no one is more is more concerned about this delay than SpeechWare, and they're doing the best they can.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make Dragon NaturallySpeaking, we make it better!

Unfortunately, patience is not a virtue unless you have it.  Equally unfortunately, few of us have it. - Unknown

-------------------------

We just got off the phone with the President of SpeechWare a few minutes ago and have received confirmation, on what appears to be, a final shipping update.

One of the future recommendations we made was to add a hyper cardioid microphone element to extend the range, as much is 6 inches, directly in front of the microphone while simultaneously reducing the pickup range from the sides and back which would result in better noise rejection while extending range. We didn't expect SpeechWare to add the improved microphone element until a future release because hyper cardioid microphone elements cost more to manufacture but SpeechWare has decided to eat that cost and replace the existing cardioid elements on all of the units before shipping them out. Even though this has caused an additional delay, in our opinion, this is excellent news because who wants to purchase yesterday's technology. Shipments out of Belgium have been re-scheduled for November 22 which now means we won't be able to ship out until the end of November or possibly as late as December 5. We probably could've gotten the 3-in-1 orders out a week or 2 earlier but as Chuck noted, we think it's worth the wait and we are already using the end product.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Can we shoot for before Dec. 31 so we can write this off????

Thanks,

Larry 

There were definitely more setbacks been we anticipated but unless something goes terribly wrong, it should definitely ship by December 5 because the shipment is already complete and on its way to Belgian. As soon as the shipment reaches Belgian, it will be parceled out and shipped directly to the US but of course it will likely be held up by customs for 24 hours. Any order over $5000 is held up in customs for at least 48 hours but we have already taken care of the paperwork and shouldn't have to deal with more than a 24-hour delay. However, we factored all of this and when we came up with our December 5 estimate.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Ship dates recede in the horizon, like a mirage.

I can't wait for it to arrive.  I'm using a Boom O and I don't know how much longer I can get away with my hair looking like this.

Thank, Lunis and all, for your attention to detail.

It is hard to wait, though.

 Vivian

Lunis,

I want to thank you and Chucker for all the efforts that you both have made to get the very best microphone available out to us who have ordered, even if it means a little more delay. Thank you all for keeping all the eager beavers informed about the reasons for the delay - try to get us the greatest mike. 

It is also very commendable that SpeechWare would replace the inner workings at their cost to get us the latest and greatest.  So I'll be Giving Thanks even tho the mike will not have arrived before then.  Happy Thanksgiving to you and all this community.

 

 

 

 

-------------------------

DNS 12.5 Pro, Win7, i7-3770K @ 3.7GHz, 16 Gb, 256 Gb SSD Boot drive, 3-in-1 TableMike - KnowBrainer 12 - Thanks so much, Lunis.

 

 

 

 

 

 

 

To the entire community:

I share these comments. Happy Thanksgiving everyone.

Gord

-------------------------

Could they please add a few additional tweaks, while they're at it?

1. Thinking. Without doubt, thinking is the rate-limiting step in my dictation process. It's hard and slow and makes my brain hurt.

2. Semantic-lexical transforms. Converting poorly formed ideas into meaningful and coherent word strings is incredibly slow for me. Yes, this is part of thinking, but a bit more concrete.

3.  Generative grammar and syntax framing. I spend far too much time working on sentences that are both well formed and effective in evoking appropriate representations in the minds of other native speakers.

OK, just 3 more tweaks for the Knowbrainer and Speechware folks, and we'll have the microphone of the century. By Thanksgiving, of course.

Your welcome, 

Matthew

 

 

Matthewls,

Love the post, very funny. Needed that today. I wish they could do as you say and just read our thoughts. I can do things like play games, move a ball with biofeedback so maybe, just maybe it is close.....

Larry 

I'm afraid those three tweaks requested by Matthewls are a little bit too much to ask at this point. Perhaps next generation. 

Thinking is hard, but having to do thinking while maintaining a particular posture in relation to the microphone is even harder.  You have to be both alive and dead at the same time, a much self-contradictory state to maintain. 

So what I would like to ask the microphone to be able to do, more realistically, is to automatically and intelligently zoom longer and shorter according to my movement. This way, at least I can remain alive when trying to do the thinking part.

But wait, that's exactly what the SpeechWare 3-in-1 has already accomplished, electronically instead of mechanically. Genius.

I think the company (or Lunis himself) should market them under a new name "i-zoom microphone", or "e-zoom microphone". Much better trademarks.

Just noticed this in the 3-in-1 description:

 NEW "De Clicking" algorithm (2nd-generation exclusive)

1.  What is this?

2.  Are those of us who have our deposits in 1st-gen?

More broadly stated, will there be any difference between the microphones that go out in the next week or 2 as part of this 1st batch vs. the microphones slated to ship Jan. 2nd?

 Thanks, John 

-------------------------

DNS 11.5 Pro / VoicePower
Work:  3-in-1 TableMike USB / Win7 64bit / i3 550 / 8GB
Home:  Sennheiser ME65/K6 / Win7 64bit

See discussion of the algorithm in 3 in 1 desk mic vs Airline 77 headset.

Fred

Thanks. 

Do we have an answer to Kent10's question yet?  (" I am wondering if they are making the change to the Hypercardioid mic permanent for all future 3-in-1 mics they sell or if this is just a bonus for the first round for those who have been waiting.  I know the hyper mic is listed as an option."

 John

 

 

-------------------------

DNS 11.5 Pro / VoicePower
Work:  3-in-1 TableMike USB / Win7 64bit / i3 550 / 8GB
Home:  Sennheiser ME65/K6 / Win7 64bit

John,

The change to the hypercardioid microphone with the reduced sensitivity (adding the 750 ohm resistor) will be permanent in all current and future SpeechWare 3-in-1 microphones. That should answer your question.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make Dragon NaturallySpeaking, we make it better!

Most of what we call management consists of making it difficult for people to get their work done. - Peter Drucker

-------------------------

Hello Lunis and Chuck and the KB forum,

 Any updates on the ETA of the 3-in-1? 

Thanks very much,

Vivian 

Vivian et al.,

Somewhere in the last 48 hours Lunis indicated that the SpeechWare 3-in-1 microphones had been shipped to him from Belgium. Understanding that these are only for those who have pre-ordered, they should be received by KnowBrainer this week, which is what he originally said in that post. So, if you have pre-ordered and you're waiting for yours, it should be shipping out shortly. However, you should check with Lunis to be absolutely certain of when they will be shipped to customers who have ordered such.

There are about three or four different posts on this topic, all containing very us bits and pieces of information. Nevertheless, Lunis is the one to really respond as to whether or not he has received the shipment and when orders will go out to customers. The most current information is that they are on the way if not here already.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make Dragon NaturallySpeaking, we make it better!

Be careful what you wish for.  You may find that what you get is not what you expect, or what you want. - Aesop (620 BC - 700 BC modern interpretation)

-------------------------

Everyone who pre-ordered their SpeechWare 3-in-1 microphone before November 28 will ship out today or tomorrow. Most ground packages will go out tomorrow, November 30, because the ground truck arrives much earlier but many ground packages shipped out today.

All non-US orders and FedEx Express orders (including FedEx 2nd Day and overnight) are shipping out today, the 29th. In other words, everyone's package is in FedEx hands.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Thank you, Lunis and Chuck!

That is great news.

I can't wait to begin using it!

Vivian 

Heigh-ho, heigh-ho!

It's time to speak we know,

We'll dictate through the whole day long,

Heigh-ho, Heigh-ho!

Good news. But Federal Express, although they say it was shipped yesterday, seems unable to track it (perhaps because they show it with two shipping dates--Oct. 3rd, the day I ordered it, and November 29th, when it shipped).


-------------------------

Dean L. Surkin

DNS 11.5, Windows 7 64-bit, Intel Q9500, 6 GB RAM

Alas, I have no news from FedEx.

 PS. Correction 12/1/2011. Lunis was so kind as to send me the fedex tracking number (that I must have deleted some time) and Voila! it's on its way, predicted to arrive tomorrow.

 THANKS LUNIS and TEAM!

Your tracking number would have been sent to you approximately 24 hours after you placed your order because we printed all of the shipping labels in advance so that we could get everyone's package out the day after they arrived. Otherwise, it could've taken up to several days for us to process the orders. Unfortunately, you can only track a FedEx number, online, for about 3 weeks and then they cease to track but you can always call FedEx (800) 463-3339 for up to the minute tracking information. We have no idea why the on-site tracking stops working after 3 weeks but that's why you don't have any information. There is no way for us to update tracking information or to send 108 tracking numbers because this would've resulted in further delays. We knew that FedEx tracking wouldn't work but we figured it was more important for our customers to receive their microphones a week early than wait for us to print shipping labels the day we send out packages.

Every 3-in-1 microphone has shipped. International and Express orders shipped yesterday while most Ground orders shipped today (about 4 hours ago). Everything is in the system and if you still have your tracking number, you can call FedEx to see exactly where it is.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Friday before 4:00pm Central time!!!!!!!!!!

Of course I am at a conference all weekend........

 Larry 

It arrived about an hour ago. Excellent packaging, build quality and instructions. I created a new profile, and will be testing it shortly.

Update (12/6/2011): I've been using it now since Friday. It is an immense improvement over the Plantronics earpiece I was using previously. I think that finally, Dragon NaturallySpeaking, together with this microphone, fulfills all of its promises. All in all, it was worth the wait and worth the money.

-------------------------

Dean L. Surkin

DNS 11.5, Windows 7 64-bit, Intel Q9500, 6 GB RAM

Mine arrived while I was at work, and now it's about 930 in the evening and I'm giving this 3-in-1 its first test. Okay, then. This transcription is happening through a virtual machine running Windows XP-64-bit allocated with 1.5 GB of virtual RAM and one virtual CPU hosted by OS X Lion 10.7.2 running on a MacBook Air 13 Inch, latest 2011 model (1.7 Ghz Intel Core I5, 4GB RAM, SSD).

I have not created a new user yet, partly because all of my microphones are piped through the host operating system to the VMware virtualized soundcard. I will later create a new user and see if it makes any difference.The speed/accuracy setting to maximum accuracy. The transcription is perfect, and relatively snappy.And this is the transcription when the speed/accuracy setting is set to maximum speed. The accuracy is also perfect, and the speed is slightly higher. So far, so good.And now, the speed accuracy trade-off is 50-50, the accuracy is still perfect, and the speed is intermediate between the first two options, but frankly, difficult really to distinguish.The most impressive part though, is that the transcription has not included an error I plug in the microphone about 60 seconds ago. Whoops, The Word "an" in the previous was transcribed as "in". I must be getting sloppy.Now the speed/accuracy slider is set to 75% accurate. Again, a slight delay occurs between the end of an utterance and the appearance of words, but the speed is so fast that the difference between this and the fastest setting is almost negligible. But, there are no errors. Perhaps this is the sweet spot, or, perhaps the microphone is in a better position.
I returned the speed/accuracy slider to 100% accurate, and I will keep it this way now, even though I'm not really waiting very long for anything and the transcription is so good.

At this point, I have to agree with Chuck and Lunis that this microphone was worth the wait.… Well, really great anyway. It seems to me now that it is slightly more accurate than the older 3-in-1. Time will tell. In any case, the fact that this microphone has its own USB soundcard and requires no batteries is a great functional improvement. My only complaint is that the LED light is so bright that I can't read the text on the button beneath I might have to put a drop of white out on the LED so it is less distracting.

Dictated into DragonPad.

Hi, Lunis. I received my SpeechWare microphone this morning, and I have it hooked up. Everything seems to be working just fine. However, I have a couple follow-up questions:

1.  in the package, there is what appears to be a second microphone, apparently to be plugged into the back of the SpeechWare base. There is a picture that seems to show this second item sitting on a table surface, plugged into the main microphone base. I don't see any reference in the instructions telling me what I should be doing with this.

 This may be a related point: When, if ever, would I want to have the indicator light be blue instead of green? 

 2.   The written materials accompanying the microphone mention some additional software.  One program is called "SpeechControl for TableMike." However, when I go to the site to download software, I get redirected to another site offering a program called "transcription aid." 

My guess is that the speech control program is not quite ready for release. Is that right?  Do you recommend the SpeechCcontrol program as a useful addition to NaturallySpeaking?

 

We believe you are referring to the included external speaker that you wear on your ear when using your microphone in the blue light setting for VoIP applications where playback is required; such as Skype. Of course you can optionally wear headphones and while we're on the subject, we tried plugging our Sennheiser HD 800 headphones into the speaker output and iTunes never sounded better. Most end-users will probably have no use for the included ear speaker. The blue light setting is for non-NaturallySpeaking Ver. 11 use.

The speech control software is for transcription work and is a separate purchase. In other words, advertising for product you probably don't need the speech recognition.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Here are two pictures of the front and back of this thing.  It looks to me like it is supposed to sit flat on a table.  I don't know whether it is supposed to be a speaker or microphone.

Your photos show the external speaker, which is to be worn over your ear.

-------------------------
DNS professional 10.1.  Windows XP. Dell latitude D620.  4GB RAM.  xTag Microphone.

As previously stated, you wear it on your ear with the ring towards the back of your ear.

PS: We know of no one who actually uses this device but it is included as a VoIP amenity.

-------------------------

Click KB 2012 REV D to Download a 30 Day Evaluation of KnowBrainer 2012 

 

 

 

Received mine today, WELL worth the wait.  After experimenting with my Sennheiser MD431 profile with the 3 in 1, I formed a new Best Match III, Orthopedic Surgery, Accuracy slider at 50% profile with it.  Very accurate, did need to train a few drug and surgical terms.  Minor item: seems to bounce out of the ADD WORD Command box as you Train a word within it, other wise seems great.  More test driving tomorrow.  Thanks to Lunis!!!!  

Mark,

Nice talking to you off line. Try your accuracy slider set to 0% and report back. I think you might find really no difference but a faster dragon.

I got mine today also even though FedEx said I would get it on Friday. That Lunis......he must of been playing a practical joke!!!!

But just got home and really tired to maybe on Sunday after my conference.

 Larry

Larry: thanks for your suggestion.  Still working with. DMPE Best Match III profile and the slider set to far left did increase speed with good accuracy.  Used it later in the afternoon and fatigue/ enunciation probably factored in, I noticed some missed technical words at the 8-10 inch distance.  My sweet spot for the 3 in 1 may be 4 to 8".  I plan to work with that over the next several days in terms of the Podiatry VOCAB vs. Ortho vs. Best Match IV etc.  Also noticed while using my mouse button  programmed to the hot key F7  for mike on/off that the mike seemed to hang up and I had to use the mouse to "manually" shut the mike off.  Maybe just a glitch in my system - older Dell Vostro  Windows XP with 4 GB RAM.  Still excited about the mike and anxious to hear other's tales of DMPE and otherwise.

G.J.,

Much ado about nothing. If I had the microphones and the 750 ohm resistors, it would take me about a minute and a half to modify each microphone.

Chuck Runquist
Technical Project Manager
VoiceTeach LLC
Home of VoicePower®: We don't make Dragon NaturallySpeaking, we make it better!

"Assumption is the mother of all screw ups" - Unknown

-------------------------

I, also, would like to thank everyone involved in this project.  We are all very excited. 

 Lunis, I am wondering if they are making the change to the Hypercardioid mic permanent for all future 3-in-1 mics they sell or if this is just a bonus for the first round for those who have been waiting.  I know the hyper mic is listed as an option.  I also wonder what the difference in price between the 2 mics would be or how much the option for the hyper mic would be.  It is not on their web site yet.  If they make the change to hyper permanent then I wonder if the price of the mic will also go up.  Thanks again!