Asked — Edited

Using Dragon Naturally Speaking With ARC

This has been brought up before is some threads, but has come up again recently. I promised that I would make a post about this so, here it goes...

There have been some posts where people have asked to be able to use Dragon Naturally Speaking (DNS) with ARC. To do this, there would have to be a separate application installed which would handle this unless there is a change to ARC itself to allow this to happen. To build an application outside of ARC to allow this would require some programming and a $2000.00 investment in the DNS developer software. To use this new tool, you would have to launch a separate application on your computer to handle speech recognition, which would pass the spoken text back to a variable in ARC. You would have to develop scripts in ARC to handle what to do with this spoken text. Another application would also need to be written that would allow the user to train DNS.

The idea of crowd funding has been brought up a few times in the past, but always in other threads that not everyone might have seen. I would like to suggest crowd funding of this idea again. It would be great if we could crowd fund the purchase of DNS Developer for EZ-Robot to add the ability to choose either DNS or Windows Speech Recognition engine directly within ARC. If that is not possible and there were enough people willing to crowd fund the purchase of DNS Developer, I would be willing to develop the 3rd party apps that could be used to handle what was described above.

This is to see how much interest there is in doing this. Please post if you would be interested in donating to this project or to donating to see if we can get this feature added directly into ARC.

One important thing to remember here. This path would only enable DNS for the applications above. It would not enable it on every application outside of these. In other words, you wouldn't have all of DNS on your machine, but only for use with ARC.


ARC Pro

Upgrade to ARC Pro

Your robot can be more than a simple automated machine with the power of ARC Pro!

United Kingdom
#1  

@David.

Great idea. Just to get the ball rolling here, this is something I would be very interested in and would be willing to donate. I very much like the idea of donating towards crowd funding the purchase of DNS Developer for the EZ-Robot team, if DJ and the guys at the office would be willing to do this for easy implementation with ARC.

You can definitely count me in for this.:)

Steve.

#2  

David, I'm all for contributing to the DNS effort, which ever direction it goes. I'm very interested in a solution.

Ed

#3  

@David, wow, you beat me to this. I was just sitting down to see if I could get this started an on track. I am willing to help with getting this started as I think it's a valuable addition for the EZ Robot platform.

As many know ARC uses the Microsoft SR engine and has no ability to use other much more advanced products like DNS. Forum members and other customers been wanting to be able to have this option for years now. However ARC is hard coded to use the MS Speech Recognition engine and it would be to much for DJ to rewrite the code (this is what I remember from past conversations a couple years ago so there may be more to it). Anyway, the only way other then waiting for MS to build a better SR engine in a future Windows release (unlikely) is to have a talented person like David write a plug-in type program that will give us the ability to choose DNS as a speech recognition engine.

I'm over the top excited that David is willing to do this. He's already written another program that works great with ARC called EZ-AI so he's proven he has the talent and know how to do this. He also has proven he's a stand up and morally based person as he has not only helped many people on this forum but has taken many hours out of his own life to personally help people "off forum" with EZ Robot problems.

EDIT: I just re-read David's first post and missed the sentence here he suggested buying this software first for EZ Robot to use in development. I'm all for that and think it's a cleaner solution. Most of what I wrote below still applies. However if EZ Robot cant go this route then the following still applies:

Why should we raise 2k for a guy we've never seen to buy personal software? This is a legitimate question. Yes, 2k is a lot of cash. Sure he will benefit in this as it would be his to do with as he sees fit. However, looking at the bigger; If he can actually provide the EZ Robot's customers and community a way to use a superior SR engine like DNS then we all benefit in huge ways. Speech Recognition is such a big part of using EZ Robot and one of the coolest features. Why not have the ability to use the best SR engine available right now?

Is this a sure thing that David can develop this plug-in and how long will it take? No, not 100%. New stuff like this is always a crap shoot and a risk. However I'm willing to bet my money on a proven, quality and knowledgeable man like David Cochran if there is a good potability of having this feature in my robot. How long will it take? There is no time line on new inventions and genius. It will take as long as it takes and we should understand that going into it. We need to let David work his magic and trust he'll do the best he can. We also need to understand that David has a real life and other responsibilities that come first.

How much should you donate? My rule of thumb in putting money into a start up without a %100 possibility of success is to only give enough money I can afford and that I'm willing to loose. With that said I'm willing to donate $200 USD at this time. No one else is expected to reveal what they are donating. We would only ask at this time to speak up to show interest in funding this.

Will you get your money back if this fails and we cant raise enough money to buy the software? We'll if this is set up correctly there should be no risk of loosing your donated funds if the software cant be bought because of lack of funding. An account could be set up through a place like Go Fund Me that will hold the cash and refund money if athe set gold is not met. I think this account should be managed by a third party member and not David. This should set people's minds to rest about the safety of the funds being raised. If people trust me, I could set this up through Go Fund Me and manage it. However, once the funds are released and the software is bought by David the money raised is non refundable. The Software id David's to do with as he wishes. However the expectation would for David to first use it to try to develop an interface between ARC and Dragon Naturally Speaking.

These were some of the questions and answers I could think of. Please ask any other ones you may think of and they will try to be answered. If there is enough interest I'm willing to help out as much as I can and do the set up work if the community and David are willing to allow me.

Thanks for our consideration. ;)

#4  

@DavidC. (or anyone wanting to chat off line about this) feel free to drop me a email at dschulpius at wi dot rr dot com. If we move forward we need to figure out the amount we should set in the funding account to raise the money and the amount needed. Remember these funding accounts require a fee usually based on the amount of money you want funded. Also after about $500 raised the funding site goes public and more then just the people you notify can see it.

#5  

I think it is a great idea to have Dave S manage the donations. I would be willing to donate around $400 to this next month. Either way (ARC or 3rd party app) is good with me personally.

If I wrote the third party app, it would be my top priority outside of my normal job and my normal life responsibilities. I would do everything in my power to deliver a working product as soon as possible. Knowing me, it would be an ongoing project to improve the capabilities as time went on as I am never satisfied with anything and never count anything as a completed project. There is always room for improvement.

If you have any questions for me, feel free to post them here. I will answer anything asked. I know that there are some others who have posted an interest in this type of solution in the past also. David, lets see how much interest there is prior to starting a go fund me. If you can find out what the cost of the fee would be for something like this, it would be helpful. I will get a quote from Dragon so that we know exactly how much would need to be raised.

[edit] I have requested a price quote for DNS SDK client. I will post the exact cost as soon as I get it back from Nuance. Thanks David

#6  

I am on the fence with this..... Maybe it would help if someone could explain what the advantage DNS has over what we have now?.... I have zero issues with speech rec now so can someone tell me what improvement or what extra I will be able to do with DNS that I can't already do with Windows speech rec?

#7  

I guess the best thing to do is to look at this

http://www.nuance.com/ucmprod/groups/dragon/@web-enus/documents/webasset/nc_008811.pdf

These are just some of the things that I see being beneficial.

DNS allows you to spell words out, for example, if facial recognition were on and a face wasn't recognized, the application could ask the user to spell their name, which could then be used to store that value in the facial recognition piece of either ARC or RoboRealm. Numbers could be recognized also. This is what Tony is using with his platform also. Dictation is more accurate too.

New in Dragon 12, developers can integrate user profile creation for an iPhone®, iPod® touch (4th gen), iPad® or compatible Android device as a remote microphone or recorder for greater freedom and flexibility.

DSC supports English (including variants for US, UK, Australian, Southeast Asian and Indian), German, French, Italian, Spanish and Dutch. Other vocabularies provided in DSC are General Vocabulary, Empty Dictation (except in Australian, Southeast Asian and Indian), and Commands Only (available in US and UK English only).

Enable continuous dictation of up to 160 words-per-minute into nearly any Windows-based application with up to 99% accuracy. Developers can customize specific requirements and leverage the full feature set of Dragon, including voice correction, formatting, and dictation playback for fast editing, as well as create new user profiles (with or without end user interaction) using Dragon’s new user wizard.

Application developers can transcribe audio files into application windows or transcribe audio files in text and document files. DSC also includes the Audio Transcribe Folder Agent, a GUI tool that automatically transcribes recorded voice from audio files into user-specified directories. This includes .wav, .wma, and .mp3 audio files, as well as .dss and .ds2 file formats.

#8  

One more thing....do you have to buy DNS or is it a free download? I am just worried that this would be a created need that I never knew I had....

Thanks David...I will do some research before I look like any more of a dumb a** than I do now...:)

United Kingdom
#9  

I just want to add to what I wrote in post #2.

It is great that David C is willing to take this on, and I cannot think of a better person for the job. His willingness to do do this, not just for his EZ-AI program, but for all of us as well, is just fantastic and I cannot thank you enough David. But whatever direction this takes, I would personally like to see some involvement from EZ-Robot with it's implementation with ARC (it is their platform after all).

DJ and the guys are striving to be the best robotics software company in the world and are currently doing a fantastic job in doing just that, which is clear to see. But to keep being the king of the castle in this field requires that the very best hardware and very best software is used, and in this case, using DNS falls in this category. And as Dave S has already said, Speech Recognition really is quite an important part of ARC and robot control in general. So I would like to think that the EZ-Robot team would be willing to take up the challenge of implementing DNS in to ARC using its members (our) donations to make this a reality and to keep ARC and EZ-Robots the best robotic software company there is, or even participate in David C's on take of this project if that's the direction this will take.

Either way, I would still be happy to donate cash and be a tester for this project whoever takes it on. My current financial situation doesn't leave me a lot of disposable income at this time, but I would be willing to donate £70 GBP (about $100 USD) to this, bearing in mind I will still need to pay out about $120 to purchase the DNS software itself, and if I can afford to pay a little more at the time of funding I would be more than happy to do just that to help make this a reality.

In any case, I am really excited about this project as I am more than a little fed up with Windows speech rec, even when I use a quality mic and lots of training. This can only be a good thing. And a quick mention to show my personal thanks to Dave S for his willingness to take care of the fund raising account.:)

Steve.

#10  

There would be no need for you to buy dragon. I will explain in a few minutes. On cell phone right now...

#11  

The reason that the DNS SDK is so much is that it allows the redistribution of the DSN components via the application that is developed. This prevents others from having to buy DNS to get these components. This is also why DNS would only work with the application that included these components (the third party app or ARC).

I would prefer that DJ took this on so that it would be in ARC, but if that isn't possible, the third party app is the other solutions that would allow this. it would be less tightly integrated with ARC but would use the hooks that are available in ARC to communicate directly with ARC via variables and port 6666 as the default. This would be configurable just like EZ-AI is configurable. It would be in the application.config file which would prevent the need for a database engine to be running.

Also, Richard R, that was a good question. I would want to know what I would stand to gain from this as well and it hadn't really been identified yet.

PRO
USA
#12  

Count me in for a donation. I would really look forward to this advancement/addition.

United Kingdom
#13  

@David.

Quote:

There would be no need for you to buy dragon.

That's good to know, and kinda makes sense to me now. Thanks for confirming. Consider my funding to go up to $200 then as this is the case;). Yeah I sure hope that DJ will jump in on this tread soon and share his thoughts on this.

A couple of thoughts to personal benefits I see of using Nuance DNS...

  1. Using ARCs speech rec works okay for me, but as my S/R command list grows ever larger, even with regular and consistent Windows V/R training, false positives and low confidence levels start to become more apparent. Also with use of the Pandorabot control using S/R and the full language library, accuracy levels are about 50% at best with a lot of my speech not being recognised at all, and that is with a good quality headset, lots of training, and mic levels set to a good level. Using my iPhone's S/R (also created by Nuance) via a VCN remote app to do this is simply a joy to use with accuracy of about 98%. I'm hoping using DNS will have similar accuracy levels for using the Pandorabot control.

  2. I don't know if the same thing stand for DNS computer software, but Nuance's iPhone recognition doesn't need user profiles to be set up, meaning that any person can pick up and use my iPhone's S/R and get great accuracy results immediately. I'm hoping DNS will be something similar which would mean anyone can talk to your robot without going through extensive Windows voice recognition training.

And the bottom line is that Nuaunce's Dragon Natural Speaking is highly rated and reviewed, and is widely used over the Windows S/R offering. To use this with ARC would be a great addition.

#14  

Like Richard R. I am undecided about this. And, like him, I have had no problem with the Windows speech engine. It's worked great right off the bat with zero training. While all the bells and whistles that DNS would bring on board are nice, the main point here is quality voice recognition. In that regard, While DNS will do a better job with the same input, will it really be enough to make all that much difference? Does anyone really know? In my view, the core issue is not so much the recognition engine as it is the quality of the audio delivered. I put my money into insuring the best quality of audio goes in first. Quality microphones and good microphone placement. Low noise amplifiers. Fast processing of the audio signal for best quality output. That sort of thing. Once those methods have been exausted, only then would I turn to better quality of recognition.

As an aside, to me, ultimately the goal is to place the audio pickup device in the robot itself. Here I would not rely on a combination speaker/microphone setup, but one, or more, small but high quality microphones with an on-board amp. At this point (for ARC anyway) the audio would be fed to ARC from the robot. This could eventually lead to an on-board processing unit that could be placed in the robot to handle the speech function. Voice processing is such an integral part of the robots we want to build after all. Nearly as integral as the ezb is for servo control. I realize placing a microphone in the robot seems like a rather round about way of doing things right now since it is the PC computer that actually does the recognition. But then, that's what we do with the camera. I also realize we have to do that with the camera in order to get the robot's perspective of what it sees, so why not also get the robot's audio perception of it's surroundings as well? Plugging the microphone into the computer is the direct route, of course, but the experience gained in talking directly to the robot will be invaluable later in the development process. It will seem more natural to people as well, while allowing for experimenting with the direction of the incoming sound. For instance the robot could turn to look at the person talking or other noises in the area. Follow a noise to find it's source. Even identify the noise by it's pattern, as we do now for the video.

#15  

The difference in the ability of dns vs ms are pretty staggering. This is why it is used in the medical field for transcription and dictation. The differences have also been tested by multiple users of this community. I personally have seen the differences as my father in law is a doctor and I have seen dns and how well it works. It is at least 2x more accurate than windows sr engine. If you do some research into what all has happened with MS speech recognition, it is easy to see why they have had the issues that they have had. This topic is really about if you would be willing to help fund the ability to choose which engine you want to use.

#16  

Fine, I get it. Too bad there is not a way to delete one's posts. All I can do is stop posting my opinions in the future.

#17  

No reasons to stop posting. You asked if anyone knew if it was better or how much better. I answered.

PRO
USA
#18  

Keep posting all questions and comments. Every one of them is relevant and helpful to all.

#19  

WBS00001, you bring up some good questions. points of view and ideas you would like to try. Please don't think your being shut down. Your point of view seems that you would rather spend your money on getting the MS speech recognition to work better then donate money to a cause that wants to try a different platform. There is nothing wrong with that point of view or even you stating it here. Good luck with your efforts and I hope you share what you do to make that happen. Lots of people including myself are interested. We want all functions of ARC to work at top efficiency and be highly accurate. With that said many of us have struggled with MS's Speech Rec used through ARC and have spent lots of money getting the fastest processors, best microphones and optimum placements of equipment only to be flustered when the robot cant even understand a simple "Thank you". Lots of people cant afford all this expensive extra equipment even if they knew what to buy or even understand how to use it. It would be great to be able to offer them and others a better option then Microsoft that is proven to work better even right out of the box. Only cost would be an extra download and some simple learning curve time with help from the community. ;)

Oh, one other thing. As far as placing the microphone on the robot. I'd love to do that also but it's been tried many times by members of this forum. There is just too much noise coming from the robot's motors and other stuff to allow this. The only thing I have been able to think would allow placement of a microphone on the robot is one that uses beam and seeking technology that will seek out and center on where a voice is coming from and cancel out other noise. This technology is just now becoming available in microphones used in conference room settings and is very expensive and large in size. Please let us know if you know of any other equipment that would work for mic placement like this. Perhaps we should start another thread exploring the different types of Speech Recognition equipment and programs that are available and what works better? :)

#20  

@d.cochran Yes, you did answer but you were snarky about it by putting in the phrase "This is not about microphone placement ... " I notice you have removed that part from your post so you must have realized that and thought better about it when I posted a response. But I also see you left in "This topic is really about if you would be willing to help fund the ability to choose which engine you want to use." This is true, still, the combination said to me, "Stick to the topic or shut up." Well, I'm the new guy here and am probably speaking out of turn and making an ass of myself (including in this post), so I'll shut up. So as to not turn this into a silly flame war (yeah, I know, too late), that is my final post in this topic.

#21  

You are right, I did remove the part about "This isn't really about microphone placement". I removed it because microphone placement does matter, however not to this particular topic. I think I have defended you on a previous post also. BTW, I removed it right after I hit Reply.

I don't get in flame battles on online forums. It is a waste of time.

The last thing that those who are interested want to see is this topic get derailed. With that said, you have made some very informative posts. I agree that being able to place the microphone in the robot would be a cool thing. I do think that having a mic port on the EZ=B would be cool. I am sure that this was considered and rejected by the EZ-B team. Nobody ever said the words that you are putting in other mouths.

The reason for the comment "This isn't really about microphone placement" was if you used DNS with a microphone in position x and used Windows SR in position x, DNS would blow the doors off of MS SR. It has been proven by many independent reviewers on the internet. I agree that there are other factors that go into good speech recognition and a good microphone and high connection speeds, along with a fast computer are all good recommendations. It makes mediocre good and good great.

The reason that I said that this is about if you want to fund the ability to have the choice of which speech recognition engine you want to use is because that is what it is about. You wont lose the ability to use the windows SR, unless DJ were to decide to do this and just replace the MS SR engine. That would be beyond our control and I don't really see him doing this, but it is possible I guess.

Please don't assume that I or anyone else is saying something that they aren't saying. It has caused more issues in community posts than I can count. I have tried to be helpful to you in the past. I am trying to be helpful to you now by providing insight into the reason that I think that this is a good project. I didn't argue any of your points and I didn't say you were wrong or say anything of a negative nature to you. Those who know me on this forum will attest that I don't do that. I am not saying that thoughts don't go through my head but I really attempt to help anyone that I can here. I am sorry if I came off short and normally spend far more time explaining my statements as I have here. I was on an Cell Phone at the time and hate typing long winded things on a cell phone.

#22  

Hi David,

i am still not finish to read the hole post ;-)

Amazing response!

What i said we will find some friends for this idea.

Ans also i am there to be sponsor this idea.

i am back again reading ;-)

Boris

#23  

Hello to all DNS friends:

Like i count we are now 6 Supporters

Steve G Dave Schulpius Richard R fxrtst WBS00001 and me Rentaprinta (Boris)

So if everybody of us can give 10% for DNS Code, we need only 4 another DNS friends.

Maybe on Monday DJ will see this post and will be also a big Sponsor, i think we can Start with this project this week.

I trust Dave to do this and also to have success!

So come on not long talking, doing!

Boris

#24  

@rentaprinta, I don't think Richard or WSB00001 have said that they will support it. I have said that I will support it at 20% though so we are still at 50% to 60% of the goal.

I do know of a couple of other people who might join in based on past conversations.

#25  

Boris don't forget Robot-Doc from the first page......

#26  

Right, so 60% to 70%. We might make it. I wouldn't be able to put in my part until July 31st. My wife and a friend started a business the other day and needed to get equipment so this month is tight.

#27  

@Robot-Doc ... sorry i did´t really know if you also want, but for this here a update:

Like i count we are still now 6 Supporters

d.cochran 20% Steve G Dave Schulpius fxrtst Robort-Doc and me Rentaprinta (Boris) 15%

So i think we are @ 75 % to the goal.

Let´s see who comes!

Boris

#28  

@david

no problem 01.08 it´s not far away!

Boo

#29  

Ya gotta love the Germans. Boris, you rock. I love your enthusiasm.

Everyone, I checked GoFundMe's fees. Here they are in Canada and the USA: GoFundMe - 5% + We Pay - 2.9% = 7.9% total.

So based on a 2K price tage we'er looking at $158 USD in fees. Since we wouldn't have to buy the DNS software I'd be willing to cover these fees along with my original $200 pledge. So the fees are not an issue now.

I think if we have a few more people say they are interested (no need to reveal how much you want to pledge) we should set up an account. We can run the fund raiser a few months and see if we can drum up the extra support. Worst that happens is we don't meet the goal and everyone gets their money back.

However we first need to see if EZ Robot is interested in this. Hopefully DJ will read this and chime in with any thoughts on EZ Robot being willing to do the work if we get the developer software for them. He may want to talk off line about this to either David C or me. DJ, If so please drop either one of us a line. There may be more going on in the background that we don't know about like copyright issues. They may be getting into legal crap that they don't want to deal with. They may not even have the resources available to work on this. eek

#30  

Definitely. DJ has first option to do this. I would love to see him say he would do it. It is an important piece of this equation. We will know more once he chimes in.

#31  

@dave

thanx for your compliment.

For the raiser fund... i hope we don´t need some or few month to get the money for the DNS licence. I hope and it must be realistic to get everything cleared until the 01.08 where David has time.

a good plan is today, not a perfect next week!

And i think DJ will be/must be interessted, because his customers (like me and you and the other guys) like this idea.

And a good firm is everytime listing to their costumers!

Boris

PRO
United Kingdom
#32  

Here is my 2 cents worth, I hope it is helpful to Richard and WBS00001 as it explains my thoughts on the adding DNS to ARC.

In my opinion adding DNS capability to ARC would be a huge plus addition for EZ-Robot!

I have been using the latest Dragon SR (DNS:13) for a year now and I can speak (dictate) a letter or document with around 99% accuracy. I first started to use Dragon (DNS:10) in 2008 in my Ai (ARIEL), here is a video of it in operation, look closely and you can see the DNS speech input box (top left corner) displaying the words that I am saying in real time.

youtu.be/cV5AJfGaEM4

The older AIMEC:3 and AIMEC:4 robots used DNS11 with approaching 100% accuracy. The EZ:1 robot currently uses Microsoft engine 8 as the robot is controlled by an EZ-B. But I still get really good accuracy with engine 8 in grammar mode (dictation mode is not so good, but grammar mode is what you use with ARC), but I use a really good wireless mic that just clips to my shirt pocket or collar. This works so well that I can even control music volumes etc as the SNR (signal to noise ratio) is so good with regards to my voice. I use the Revolabs xTAG its pricey but you have to pay if you want this level of performance.

www.revolabs.com/products/product-line/xtag-usb

Note - grammar mode is much easier for the SR engine as it only has to compare incoming speech with existing grammar files - in dictation mode you can say anything and it does not require pre-programmed grammar files.

I would also happily donate to this great cause, so please count me in.

Tony

United Kingdom
#33  

As others have said, DNS is far superior to MS, David C said about 2 times, I would say it's much higher than that.

Accuracy in dictation out of the box is insanely high, MS on the other hand is ridiculously low.

Grammar mode, MS works OK. I use MS and Grammar mode with "payload" files which hold the relevant words or phrases for the command I am speaking, for instance if I ask it to "Play music by <artist>" it will recognise the Play music by part then check the <artist> as a separate payload. The payload files do dynamically update (however accuracy is still low on this, even after 5 years of training).

5 years ago I would have jumped on this knowing what I know now. Now, after 5 years of development and training it's not really worth my while but anyone who hasn't spent 5 years of constantly training MS voice recognition (or just wants a very accurate dictation mode) it's a no brainer to use DNS over MS.

#35  

If I wasn't financially tapped out due to two back to back plumbing disasters in my home, I would donate to this.

What i can do is talk to my Nuance sales rep and see if I can get us a discount on the SDK (no guarantees). I work with Nuance extensively with their enterprise VRU product. My employer spends millions of dollars with them every year, but it is a different division of the company, so I don't know how much my rep could help.

Alan

#36  

I would be interested in donating too but that would depend on the amount. Retired here so you can understand that.

Canada
#37  

I'm in too, only if the final product can support french language.

#38  

Hey Aerius,

This is from their document on the developer kit.

DSC supports English (including variants for US, UK, Australian, Southeast Asian and Indian), German, French, Italian, Spanish and Dutch. Other vocabularies provided in DSC are General Vocabulary, Empty Dictation (except in Australian, Southeast Asian and Indian), and Commands Only (available in US and UK English only).

#39  

David S,

I think that there is sufficient interest in going forward. Lets setup a go fund me account and have it run for a couple of months. If the money is available early, we can start early. During that time, we can see which way we need to go as far as if DJ would be able to add this into ARC, or if a 3rd party app needs to be developed.

I think all are in agreement that having EZ-Robot do this would be the best option. The 3rd party app will be the fallback incase this isn't doable now by the EZ-Robot team.

Thanks David

#40  

@TheTechGuru,

If you could see if it is possible to get a discount, that would be awesome. Please let us know what you find out.

Thanks man. We all appreciate it.

#41  

Question has been asked. I got an out-of-office reply that she is on vacation for independence day through tomorrow, so Wednesday would be the earliest we can expect an answer, and I told her it was low priority since it is not a query for my employer.

Alan

#43  

I think this is a very worthwhile project for the EZ community and am willing to make a donation.

#44  

When the account gets set up I'll throw in a few dollars.

PRO
Synthiam
#45  

I will schedule some time this week to do a thorough review of DNS and their API etc.. It might also be useful if EZ-Robot contacts DNS to see if they have a licensing program for startups.

#47  

Ya baby! I'm in heaven. Thanks for having our backs DJ!

United Kingdom
#49  

@DJ.

Just to echo what David C, Dave S, and Will have just said, it's great to hear from you about this project and thanks in for looking in to this further. This would be a truly great addition to the already wonderful platform you have created.:)

#50  

We will give DJ time to review DNS's API and make a determination. After that point we will setup the GoFundMe account and word the description of the account as either Buy for EZ-Robot or third party app. David S will post the GoFundMe account after this determination has been made.

United Kingdom
#51  

Sounds good to me. Cheers David.

#52  

@david

you see the people love the idea and i believe we have fast success with this Project and all our Robots can understand and speak diffrent languages better in the future.

@dave

So please check up this raiser fund konto

@DJ

nict to see you here!

@all

And hello to the new friends of the DNS Project

Supporter UpdateV2 (12 Supporter):

d.cochran Steve G Dave Schulpius fxrtst Robort-Doc OldBotBuilder Zap Brannigan Toymaker Herr Ball Aerius merne and me Rentaprinta (Boris)

#53  

I'm back. You can count me in fund the project if needed. At David thanks for bringing this subject up I think it's a great idea. I have 100% confidence if you took on the project you would complete it you have help me out in the past with many hours and asking for nothing in return. thank you .

Thank you DJ and ez-robot team

#54  

Mike, great to see you again. Thanks for supporting this. Hope everything is good. Missed having you on here.

#55  

Hello everbody,

First: i want to move this article to the top of the forum.

Second: I want to ask if there is any update:

@DJ : Do you check something with the API of DNS?

@thetechguru : Do you get any answer from your DNS Sales rep?

@d.cochran : Hi David, any news from DNS?

Boris

#56  

I haven't heard anything back from them, although I do have an update.

There is another product that I am starting once I can get my hands on Windows 10 and on Developer Studio 2015. I met with my investor to finalize some things last night. I will be purchasing DNS Developer and some other things for use with a new product that uses a lot of the EZ-AI components and some new things that I will develop. This software won't be free but EZ-AI up until version 1.0.0.8 will be free. This will incorporate some other API's that also cost money to use so we would have to charge for the use of this product. One of these is the Watson API's that were mentioned in another thread.

The agreement is that any product that uses controls that were purchased by my investor will be under the control of Cochran Robotics Inc GP, which he is a 50% partner of. I did discuss with him about having multiple versions (a free one and a paid one). The free one would demonstrate the capabilities of the product for either a limited number of transactions, or a time frame, but would allow other things that were already developed by me to be available for use for free. He was in agreement with this. It still remains to be decided what the final plan on this will be.

In any event, I will be purchasing DNS Developer and will be able to either assist or develop something that could be used with ARC. It looks like the timeframe will be sometime next month. I need to document all of the things that a new product will be able to do during the remainder of this month and come up with a look/feel of the product so, that will be my focus over the next couple of weeks. The good news though is that funding will be available to take a product from a concept through production without needed to be crowd funded.

With all of this being said, I do still strongly believe we should crowd fund the purchase of DNS Developer for EZ-Robot.

United Kingdom
#57  

Thanks for the update David. It sounds like you've got big and exciting things planned for EZ-AI. Great to hear and wish you luck with it. I'll be watching this space for your progress, and to see what Alan and DJ find out in regards to DNS.

#58  

So, first, congrats on the exciting progress with your start up business. Your talents deserve rewards and I'm sure you will succeed.

Second, for clarity; What I read you saying is that any product coming from you that would use the DNS through ARC like the 3rd party app we were discussing will now have to charge to use. Am I mis-reading this? If this is correct then the only option we now have for DNS support for no charge is if DJ decides to incorporate it nativity into ARC. I also note you are willing to assist DJ with development if he asks you.

If this is correct I'm not sure how to feel about your decision. I guess I cant blame you and your partner to try to make a product that would be in demand and will raise money and support to push it forward. However on the other hand I feel kinda abandoned and turned into a revenue source. If DJ decides he doesn't want to move forward we're screwed unless we buy from you.

Please correct me if I've mis-understood your new position. I'll gladly edit my comments to remove any concerns or qustions about your motives. :)

#59  

Hey David,

So here are my thoughts. There would be a use for multiple copies of dns developer as I will be bringing on another developer. If crowd funding happened and DJ were to decide to not add it to ARC, I could use this other licence (crowd funded one) to develop the application (ARC to DNS bridge) we had discussed earlier without making it property of cochran robotics, and thus avoid any issues that I might run into.

The same offer still stands. I am willing to develop a tool that could be used with ARC for DNS. I would just be careful to avoid any issues with legalities and all.

#60  

Btw, this would be free to use as it wouldn't fall under the partnership and would be developed using tools not purchased by the partnership.

No need to edit post. These are very valid concerns.

#61  

One more thing. I do believe that DJ would add this to ARC because it expands the features of this product especially where IOS and Android are concerned, but also the windows based ARC. I don't think that I would be the one developing this and I do believe that it will be purchased for ez-robot ultimately. This is the path that has the best logical outcome. I haven't had any discussions with DJ that makes me believe this. It is only based on what I believe he will do.

#62  

@DavidC. Understood and I respectfully thank you for making this clear enough for me to understand. Thank you for standing by your word. That shows integrity, good moral strength and direction. This is huge and becoming more rare in today's world. I'll continue to support you and EZ Robot any way I can. ;)

I wish I was as sure as you are that DJ will support DNS. He's a busy guy with busy people working for him. There are a lot of roadblocks I can see that would make it hard for him to implement this.

United Kingdom
#63  

@David C.

I just wanted to echo what Dave S just said. Fair play to you sir.

PRO
USA
#64  

David,

Do you have a time line as to the evolvement of EZ-AI? I think I understand that EZ-AI is a part of the whole with products you are developing with this investor. Will it take a back burner for now and developed later as a paid product?

#65  

Hi David,

it sounds really good what you are doing.

I have no problem with a paid version, if it is not so expensive!

It also the question what this new software product can do.

Do you heard anything from DJ?

Boris

#66  

Sorry, been rewiring lights on a trailer.

@boris, I wont talk about the other product right now other than to say that it is software that is a part of something that we are manufacturing and will be selling. Sorry, I have to protect my interests in this area. I hope you understand.

@Will, The new software will be written using UWP. EZ-AI remains mine. parts of EZ-AI will be used in the other software (even maybe all of it). If there were a community funded purchase of DNS Dev, I would be able to continue development to improve the SR with EZ-AI. If not, I would just leave the SR as it is and continue development in time.

EZ-AI will remain a classic app. The new product will have software written and rewritten for it that is a uwp app. This allows me to differentiate what is owned by me and what is owned by Cochran Robotics. If there is something added to the other app that I think would be cool in EZ-AI, I will add it.

Legalities and partnerships make things more complicated for sure, but I think it will benefit both products in the long run. Anytime you write something once, you write it the second time far better. Who knows which will be worked on first vs second, but because I have a timeline in my head as to when we need to get things to market, I have a feeling that EZ-AI modifications would be second, which is a good and bad thing.

#67  

EZAI will never be a paid product, but some features of the product may require that some sort of billing is setup depending on if there are fees associated from the vendors of the apis for usage.

You would only be charged something if you used a feature (which will be identified) that required a charge. It might be that I don't include my api key for those items and require the user to get a key, which would have them then setup billing with that vendor.

#68  

Hi David,

no prob, i understand!

In the Moment i am more interessted in the DNS App.

But i hope you are also have success with your new software!

Boris

#69  

I sent another request for a quote, this time to the correct sales department. I'll let you know if I hear back from them.

#70  

After having multiple emails ignored, and then having multiple phone calls dropped before I could contact someone, I decided to see if there was a reseller locally that carried the product. I just got off the phone. He is supposed to be getting pricing information for me. I will pass this along here when and if I hear back from him.

United Kingdom
#71  

Their customer service sounds a bit... well, rubbish. Let's see if you have some better luck with the re-seller.

#72  

Well, that was a dead end...

Here is the email I got back from him.

Hello Mr. Cochran. Thank you very much for your call regarding Dragon SDK. As a reseller of Nuance products and being certified as a trainer for NaturallySpeaking it was difficult to narrow down information regarding the SDK. I have made a few phone calls and in the meantime here is a link to a data sheet:

 

http://www.nuance.com/ucmprod/groups/dragon/@web-enus/documents/webasset/nc_008811.pdf

 

Also, this is a link to a fillable form requesting a callback for Dragon SDK.

 

http://www.nuance.com/dragon/contact-sdk-sales/index.htm

 

My apologies for not being able to supply you with more information, if some of my contacts to get back to me I will certainly pass on what I can.

 

Krysler Consulting is a value-added reseller of Nuance products, primarily NaturallySpeaking professional, legal, medical and premium. We have remote access capabilities and we have a certified trainer on staff. If you have any questions I will do my best to help you in any way I can.

 

Thank you very much for your call.

#73  

I'll give my sales rep another poke. She sells enterprise services to federal accounts so unfortunately not the department we need, but I am sure she can get me a name and number at least.

Alan

#74  

Thanks Alan. This guy is going to speak to someone he knows there that used to work in sdk sales support. Working it from 2 angles is probably the best.

#75  

My rep got back to me this morning, she has been catching up on things from vacation so was delayed. She is going to track down a rep in the right department and get back to us.

Alan

United Kingdom
#77  

Nice one. Thanks Alan. Fingers crossed.

#78  

Cool Alan!

I hope she comes back with good news!

Boris

Brazil
#79  

count me in! just remind me how and how much do I need to contribute. Greetings from Brazil!

#80  

Hi Brazil!

Greetings back from cold Germany. ;-)

Boris

#81  

I heard from Nuanve Sales (David and DJ, I'll forward you the contact information).

Bad news, the current license is not the structure we thought it was. It is $5000 per developer and $324 per speaker, so not free distribution in your app. However, they have a totally free web based platform (similar to the PandoraBots model) that we could utilize.

I am on the train replying by phone, but when I am in front of my computer later I'll send the details.

Alan

#82  

First, thank you for the information.

The web platform costs per transaction after 1000 uses I think. The $324 per speaker is what my main fear was.

I can get my company to fund 1/2 of the developer license but, that doesn't resolve the $324 per speaker issue.

I will post a link to their cloud solution.

#84  

The web is free for 1 million transactions per month, then is 8/10 of a cent per use. (per application, would need to dig into the docs to see if that is per instance, or all users, or if each user could sign up individually so the transactions aren't aggregate across all users).

I see the rep copied you on his response. I'll forward to DJ.

Note, although they don't advertise it, this is the same platform that runs iPhone's SIRI, as well as many companies IVRs (Amtrak, Fedex, many ohers).

#85  

I missed the gold level required for REST. Hmm, how do we handle $. 008 per transaction across multiple users. More research required.

Alan

#86  

Yes, that is the issue. I could develope something to use this but then each user would have to go in and register to use it with Nuance. I could have the key codes and other things required in a config file so that billing would just happen through them. Do you guys want me to give that a shot?

#87  

I think that makes sense, especially since they take PayPal. Each user could just pay for what they use.

Alan

#88  

I have setup an account. Waiting to see what all comes back. Seems like I wait a lot on them:)

United Kingdom
#89  

Yeah, sounds good to me David. And thanks for the effort you and Alan are putting in to the research and updates.

Very greatful for your help indeed.:)

#90  

Watson is another option for this. Speech to text rates are as follows.

First thousand minutes per month are FREE. Additional minutes are $0.02 per minute.

#91  

I will get setup on both of these services. I need to figure out what I am going to use anyway so, this will be a good chance to test out both options.

#92  

I'll be interested in seeing how Watson performs. 5 years ago I was evaluating platfirms that used IBM, Nuance, and Microsoft (not SAPI, an enterprise product they wound up selling). Nuance was far superior, but IBM has had 5 years to catch up since I saw them. In my more recent evaluation, none of the vendors used IBM or Microsoft. Most used Nuance.

Alan

#93  

I have just tested ibm and it worked better than Microsoft in dictation mode. It actually did pretty well. I have requested the dns cloud dev info but haven't gotten an email back yet. I do have to say that I was very impressed with how well watson stayed up with spoken text and you could see it changing other words already spoken based on the words new context of the words. Pretty cool.

#94  

Hi,

i check this all youtube videos about watson.

Amazing!

Specialy the "Watson Jeopardy" ;-)

Do you know if Watson is also multilanguage?

Boris

#97  

Right now Speech to Text in Watson is only in these languages English (US), Japanese, Spanish

There is a translator though but translated text doesn't sound right most of the time, but the translator converts these languages French, Spanish, Portuguese, Arabic and English

Sorry, no German. I will have to look at what the DNS Cloud app can convert or recognize. I know it was pretty extensive.

#98  

DNS has a lot of languages in their cloud app.

http://dragonmobile.nuancemobiledeveloper.com/public/index.php?task=supportedLanguages

This site would show you the possibilities.

Speech to text Arabic (Egypt) ar_EG ara-EGY 8 kHz, 16 kHz Arabic (Saudi Arabia) ar_SA ara-SAU 8 kHz, 16 kHz Arabic (International) ar_AE ara-XWW 8 kHz, 16 kHz Bahasa (Indonesia) id_ID ind-IDN 8 kHz, 16 kHz Cantonese (Simplified) zh_HK yue-CHN 8 kHz, 16 kHz Catalan ca_ES cat-ESP 8 kHz, 16 kHz Croatian N/A hrv-HRV 8 kHz, 16 kHz Czech cs_CZ ces-CZE 8 kHz, 16 kHz Danish da_DK dan-DNK 8 kHz, 16 kHz Dutch nl_NL nld-NLD 8 kHz, 16 kHz English (Australia)* en_AU eng-AUS 8 kHz, 16 kHz English (GB)* en_GB eng-GBR 8 kHz, 16 kHz English (US)* en_US eng-USA 8 kHz, 16 kHz English (India) N/A eng-IND 8 kHz, 16 kHz Finnish fi_FI fin-FIN 8 kHz, 16 kHz French (Canada) fr_CA fra-CAN 8 kHz, 16 kHz French (France)* fr_FR fra-FRA 8 kHz, 16 kHz German* de_DE deu-DEU 8 kHz, 16 kHz Greek el_GR ell-GRC 8 kHz, 16 kHz Hebrew he_IL heb-ISR 8 kHz, 16 kHz Hindi hi_IN hin-IND 8 kHz, 16 kHz Hungarian hu_HU hun-HUN 8 kHz, 16 kHz Italian it_IT ita-ITA 8 kHz, 16 kHz Japanese ja_JP jpn-JPN 8 kHz, 16 kHz Korean ko_KR kor-KOR 8 kHz, 16 kHz Malay ms_MY zlm-MYS 8 kHz, 16 kHz Mandarin (China / Simplified) cn_MA cmn-CHN 8 kHz, 16 kHz Mandarin (Taiwan / Traditional) zh_TW cmn-TWN 8 kHz, 16 kHz Norwegian no_NO nor-NOR 8 kHz, 16 kHz Polish pl_PL pol-POL 8 kHz, 16 kHz Portuguese (Brazil) pt_BR por-BRA 8 kHz, 16 kHz Portuguese (Portugal) pt_PT por-PRT 8 kHz, 16 kHz Romanian ro_RO ron-ROU 8 kHz, 16 kHz Russian ru_RU rus-RUS 8 kHz, 16 kHz Slovak sk_SK slk-SVK 8 kHz, 16 kHz Spanish (Spain) es_ES spa-ESP 8 kHz, 16 kHz Spanish (LatAm) es_MX spa-XLA 8 kHz, 16 kHz Swedish sv_SE swe-SWE 8 kHz, 16 kHz Thai th_TH tha-THA 8 kHz, 16 kHz Turkish tr_TR tur-TUR 8 kHz, 16 kHz Ukrainian uk_UA ukr-UKR 8 kHz, 16 kHz Vietnamese vi_VN vie-VNM 8 kHz, 16 kHz

Text to speach Language 4 char * 6 char * Voice Gender Arabic ar_WW ara-XWW Laila F Arabic ar_WW ara-XWW Maged M Arabic ar_WW ara-XWW Tarik M Bahasa (Indonesia) id_ID ind-IDN Damayanti F Basque eu_ES baq-ESP Miren F Cantonese zh_HK yue-CHN Sin-Ji F Catalan ca_ES cat-ESP Jordi M Catalan ca_ES cat-ESP Montserrat F Czech cs_CZ ces-CZE Iveta F Czech cs_CZ ces-CZE Zuzana F Danish da_DK dan-DNK Ida F Danish da_DK dan-DNK Magnus M Dutch nl_NL nld-NLD Claire F Dutch nl_NL nld-NLD Xander M Dutch (Belgium) nl_BE nld-BEL Ellen F English (Australia) en_AU eng-AUS Karen F English (Australia) en_AU eng-AUS Lee M English (GB) en_UK eng-GBR Kate F English (GB) en_UK eng-GBR Serena F English (GB) en_UK eng-GBR Daniel M English (GB) en_UK eng-GBR Oliver M English (India) en_IN eng-IND Veena F English (Ireland) en_IE eng-IRL Moira F English (Scotland) en_SC eng-SCT Fiona F English(South Africa) en_ZA eng-ZAF Tessa F English (US) en_US eng-USA Ava F English (US) en_US eng-USA Allison F English (US) en_US eng-USA Samantha F English (US) en_US eng-USA Susan F English (US) en_US eng-USA Zoe F English (US) en_US eng-USA Tom M Finnish fi_FI fin-FIN Satu F French fr_FR fra-FRA Audrey-ML F French fr_FR fra-FRA Thomas M French fr_FR fra-FRA Aurelie F French (Canada) fr_CA fra-CAN Amelie F French (Canada) fr_CA fra-CAN Chantal F French (Canada) fr_CA fra-CAN Nicolas M Galician gl_ES glg-ESP Carmela F German de_DE deu-DEU Anna-ML F German de_DE deu-DEU Petra-ML F German de_DE deu-DEU Markus M German de_DE deu-DEU Yannick M Greek el_GR ell-GRC Melina F Greek el_GR ell-GRC Nikos M Hebrew he_IL heb-ISR Carmit F Hindi hi_IN hin-IND Lekha F Hungarian hu_HU hun-HUN Mariska F Italian it_IT ita-ITA Alice-ML F Italian it_IT ita-ITA Federica F Italian it_IT ita-ITA Paola F Italian it_IT ita-ITA Luca M Japanese jp_JP jpn-JPN Kyoko F Japanese jp_JP jpn-JPN Otoya M Korean ko_KR kor-KOR Sora F Mandarin (China) zh_CN cmn-CHN Tian-Tian F Mandarin (Taiwan) zh_TW cmn-TWN Mei-Jia F Norwegian no_NO nor-NOR Nora F Norwegian no_NO nor-NOR Henrik M Polish pl_PL pol-POL Ewa F Polish pl_PL pol-POL Zosia F Portuguese (Brazil) pt_BR por-BRA Luciana F Portuguese (Brazil) pt_BR por-BRA Felipe M Portuguese(Portugal) pt_PT por-PRT Catarina F Portuguese(Portugal) pt_PT por-PRT Joana F Romanian ro_RO ron-ROU Ioana F Russian ru_RU rus-RUS Katya F Russian ru_RU rus-RUS Milena F Russian ru_RU rus-RUS Yuri M Slovak sk_SK slk-SVK Laura F Spanish (Castilian) es_ES spa-ESP Monica F Spanish (Castilian) es_ES spa-ESP Jorge M Spanish (Columbia) es_CO spa-COL Soledad F Spanish (Columbia) es_CO spa-COL Carlos M Spanish (Mexico) es_MX spa-MEX Angelica F Spanish (Mexico) es_MX spa-MEX Paulina F Spanish (Mexico) es_MX spa-MEX Juan M Swedish sv_SE swe-SWE Alva F Swedish sv_SE swe-SWE Oskar M Thai th_TH tha-THA Kanya F Turkish tr_TR tur-TUR Cem M Turkish tr_TR tur-TUR Yelda F Valencian es_ES spa-ESP Empar F

#99  

@DJ Sures,

I don't have your direct email address (I could guess, but I suspect you get so much email it would get buried anyway), so I forwarded the Nuance contact information to customersupport@ez-robot.com with a request to make sure you see it. If you want, you can contact me directly. Email in my profile.

Alan

PRO
Synthiam
#100  

awesome - that will go to aislinn. I have it on my list and was hoping to get at it this week.

Germany
#101  

@cochran Robotics

Please give me more information for your EZ-AI Server. When I buy this, then does the language control and the language output in German or only in English? I'm looking for a way to include a chatbot in ARC. Unfortunately this does not work properly with Pandora, since the Windows language recognition is too bad. I also use Dragon naturally Speaking, but I do not know how I should integrate in ARC. Would EZ-AI be a solution? Thank you very much

Sven