8 hours ago, kevinfirth said:

I have many of the ivona voices

Cereproc & Nuaunce (VE) as well, over 20 ..... and any of which can be your copilot in both Vox & MCE.

A shared investment you can say. And then to hear all calsigns, paired number flite #'s, AI comms, pronounced waypoint & airports, a lot better/real than missing and / or phonetics.

Edited April 14, 20215 yr by vadriver

This sounds interesting, no pun intended, so if you buy multiple voices can they be used simultaneously, different voices for different aircraft and controllers?

33 minutes ago, jon b said:

This sounds interesting, no pun intended, so if you buy multiple voices can they be used simultaneously, different voices for different aircraft and controllers?

Yes, here's how Voxatc voice config page looks like with many addon voices.  You can select pilot and controller voices, filtered, with static or as provided.  Gives a much more authentic experiences with a lot of variety.  I picked up all my Ivona and Cereproc SAPI voices when they had sales on.

Edited April 15, 20215 yr by kevinfirth

all I red from this thread is there hasn't been a newly develop ATC software in a long time. They all look like they were done in the 80s

23 minutes ago, fogboundturtle said:

all I red from this thread is there hasn't been a newly develop ATC software in a long time. They all look like they were done in the 80s

Some are still around but not unsupported, others like VoxATC are still being updated and improved.  

There haven't been any completely new ATC apps I agree, but that's because its fiendishly difficult to do and its a niche product for a niche product, so tends to be commercially unviable in terms of them returning much profit to developers.

Just now, kevinfirth said:

Some are still around but not unsupported, others like VoxATC are still being updated and improved.  

There haven't been any completely new ATC apps I agree, but that's because its fiendishly difficult to do and its a niche product for a niche product, so tends to be commercially unviable in terms of them returning much profit to developers.

If I was younger I would use build a brand new ATC using Machine Learning. I would collect all the date from VATSIM /PilotEdge controller and let the AI make decision on how to route all of the traffic. It would be exactly like the real thing for the most part. 

On 4/14/2021 at 11:12 AM, kevinfirth said:

Never seen that here, the only way I can see that happening is if default AI is active at the same time (which it should never be!)

here's an example though of what I mean, it may not be typical but I've never seen an ATC control ground and air at the same time. The example starts at 1:13

 

 

1 hour ago, FPStewy said:

here's an example though of what I mean, it may not be typical but I've never seen an ATC control ground and air at the same time. The example starts at 1:13

 

 

That is indeed strange, I noticed if I have the traffic levels too high Vox can’t handle it sometimes anymore.

The best is of course Vatsim, but for me with two small children not allways possible.

 

1 hour ago, FPStewy said:

I've never seen an ATC control ground and air at the same time.

Actually, VOXATC is simultaneously communicating with all AI traffic. If you set com2 to one of those frequencies you will hear it. In full debug mode VOXATC creates text files (.log) for the comms for each aircraft and each radio frequency. Somewhere in this subforum there's a post of how to display these log files dynamically will running VOXATC.

1 hour ago, fogboundturtle said:

If I was younger I would use build a brand new ATC using Machine Learning. I would collect all the date from VATSIM /PilotEdge controller and let the AI make decision on how to route all of the traffic. It would be exactly like the real thing for the most part. 

I’m not sure what machine learning involves. Do you have a degree in mathematics? That is the level of knowledge required to determine not only your position in the flight sim world but also every Ai aircraft.

I can remember when Radar Contact v4 was being developed and as a member of the alpha & beta team the complexity of the trigonometry required was breathtaking. We had a maths genius on the team who was extremely helpful.

That explains why ATC programs are rarer than hen’s teeth. Probably the most complex addon to a flight sim program.

2 hours ago, FPStewy said:

here's an example though of what I mean, it may not be typical but I've never seen an ATC control ground and air at the same time. The example starts at 1:13

 

 

As @jabloomf1230 says vox continually communicates with AI.  Ground controllers and approach are on separate freqs for a reason, so it's not surprising you've not seen an ATC station control both ground and airborne traffic at the same time.  The minor exception to this might be a/c that are handed over to the tower from approach and from ground while a/c are operating on or near the runway.

58 minutes ago, Ray Proudfoot said:

I’m not sure what machine learning involves. Do you have a degree in mathematics? That is the level of knowledge required to determine not only your position in the flight sim world but also every Ai aircraft.

I can remember when Radar Contact v4 was being developed and as a member of the alpha & beta team the complexity of the trigonometry required was breathtaking. We had a maths genius on the team who was extremely helpful.

That explains why ATC programs are rarer than hen’s teeth. Probably the most complex addon to a flight sim program.

Yes you need to have a good knowledge base of mathematics to write the equations after that is all machine learning. The more data your feed to the engine, the better it gets at making decision. This is a few Machine Learning formulas

https://www.datasciencecentral.com/profiles/blogs/140-machine-learning-formulas

Another interesting article : https://www.hindawi.com/journals/complexity/2020/1659103/

Edited April 15, 20215 yr by fogboundturtle

21 minutes ago, fogboundturtle said:

Yes you need to have a good knowledge base of mathematics to write the equations after that is all machine learning. The more data your feed to the engine, the better it gets at making decision. This is a few Machine Learning formulas

https://www.datasciencecentral.com/profiles/blogs/140-machine-learning-formulas

Another interesting article : https://www.hindawi.com/journals/complexity/2020/1659103/

I'm sorry but that means nothing to me. John Dekker wrote RC4 using Visual Basic which seem okay to me despite the enormous increase in Ai injected into flight sim compared to 15 years ago.

Do you have any experience of writing programs that interact with P3D etc?

Oh, one other thing. We had two real-world air traffic controllers on the team. One for FAA airspace and the other for ICAO. You'll need access to someone with those skills to get the rules and phraseology correct.

5 minutes ago, Ray Proudfoot said:

I'm sorry but that means nothing to me. John Dekker wrote RC4 using Visual Basic which seem okay to me despite the enormous increase in Ai injected into flight sim compared to 15 years ago.

Do you have any experience of writing programs that interact with P3D etc?

Oh, one other thing. We had two real-world air traffic controllers on the team. One for FAA airspace and the other for ICAO. You'll need access to someone with those skills to get the rules and phraseology correct.

Like I said, I am too old for that now. I wouldn't want to start a project. 

I can try to explain to you in laments terms how Machine Learning works.  For each event that requires and action, You feed the equation the right answer.  You need to feed the equation millions of transaction that tells the AI what is the right answer. What the machine learning does is approximate the right answer.  The more data it has, the better it gets at predicting the right decision. So it's not your typical If <plane does this> visual basic logic.  The AI will determine how to control all the traffic in it's region based on what it learned.

The rules of ATC communication are myriad but they haven't changed that much over time. Since they are all written down somewhere, it just takes perseverance to code an AI ATC app, not any fancy neural network computer, like which might used in designing the algorithms for self driving cars. The former is just a massive collection of IF and CASE statements. The latter has to process sensor data in real time. Is that an object? Is it a dog? How  does it move? Etc.