The new voice codec

[Klaus Schmitzer]

Hello!

 

Cant hide my impatience towards the new voice codec

Is there any news or release info available out there?

 

Thank you very much!

[Nicholas Cavacini]

From what I heard, this is still in the very early planning stages.

[Camden Bruno]

Klaus,

 

When there are updates, I'm sure they'll be posted. Patience, young padawan.

[Matthew Cianfarani]

Klaus,

 

From the announcement (https://www.vatsim.net/news/new-voice-codec-voice-unicom)

 

At Connexion 2017, Kieran Hardern, who sits on the VATSIM Board of Governors, made an announcement that VATSIM has a team actively working on the implementation of a new voice codec. A number of tests of specific codecs have already been completed and VATSIM is committed to make this happen. As with any voluntary project, we won't be putting an estimated time of release out on this but hope this announcement is of interest to our members who have been eagerly waiting for this upgrade.

[Klaus Schmitzer]

Thank you guys! Normaly I am an very patient person, but this will bring me back to vatsim, so I cant hold back

 

Okay, I will sit back and wait!

[Camden Bruno]

this will bring me back to vatsim, so I cant hold back

I can't hold back either: You're not participating on the network just because of the codec..?

[Don Desfosse]

Yeah, nothing in life is perfect.... Best to keep participating on the network looking for improvement, rather than staying away waiting for an improvement.

[Klaus Schmitzer]

I never flew much online because of too much garbeled transmissions. I know, you alwyas can say "please repeat" and so on... but this breaks all learning (at least for me, not talking for others here) most voice on Live ATC is way better to understand. So I never gathered much experience.

[Robert Cardone 1323915]

I never flew much online because of too much garbeled transmissions. I know, you alwyas can say "please repeat" and so on... but this breaks all learning (at least for me, not talking for others here) most voice on Live ATC is way better to understand. So I never gathered much experience.

 

As a retired broadcast engineer, I think the garbled audio is more to do with who is speaking. Sometimes the audio sounds like the person is in another room, probably because they are not close to the mic, or the mic is a piece of junk, or they never bothered to calibrate their audio levels. Add to that, some people just mumble instead of speaking clearly. Tonight I flew in a session, and most pilots had very clear and understandable audio. A few, you couldn't make out what they were saying at all. That is not a codec problem.

[Klaus Schmitzer]

I never flew much online because of too much garbeled transmissions. I know, you alwyas can say "please repeat" and so on... but this breaks all learning (at least for me, not talking for others here) most voice on Live ATC is way better to understand. So I never gathered much experience.

 

As a retired broadcast engineer, I think the garbled audio is more to do with who is speaking. Sometimes the audio sounds like the person is in another room, probably because they are not close to the mic, or the mic is a piece of junk, or they never bothered to calibrate their audio levels. Add to that, some people just mumble instead of speaking clearly. Tonight I flew in a session, and most pilots had very clear and understandable audio. A few, you couldn't make out what they were saying at all. That is not a codec problem.

 

I absolute agree with you, but I think it helps a lot if the transmission itself works better - even if someone is mumbling.

[Camden Bruno]

Live ATC is way better to understand.

I don't know which Live ATC streams you're listening to, but do you have any real-world flying experience? If so, I'd be willing to bet that you'd appreciate the fact that our voice codec is actually extremely realistic. Apart from the delay, the codec does a nice job of mimicking the frustrations, garbles, weird noises, etc. that we constantly hear on real-world frequencies. Having a voice codec like PilotEdge, where it sounds like you're talking on Skype or TeamSpeak, takes away from the experience in my opinion.

 

To each their own.

 

Take care,

[Robert Cardone 1323915]

Live ATC is way better to understand.

I don't know which Live ATC streams you're listening to, but do you have any real-world flying experience? If so, I'd be willing to bet that you'd appreciate the fact that our voice codec is actually extremely realistic. Apart from the delay, the codec does a nice job of mimicking the frustrations, garbles, weird noises, etc. that we constantly hear on real-world frequencies. Having a voice codec like PilotEdge, where it sounds like you're talking on Skype or TeamSpeak, takes away from the experience in my opinion.

 

To each their own.

 

Take care,

 

Totally agree. I tried Pilotedge and I felt like I was talking to someone in a TV studio.

[Klaus Schmitzer]

Live ATC is way better to understand.

I don't know which Live ATC streams you're listening to, but do you have any real-world flying experience? If so, I'd be willing to bet that you'd appreciate the fact that our voice codec is actually extremely realistic. Apart from the delay, the codec does a nice job of mimicking the frustrations, garbles, weird noises, etc. that we constantly hear on real-world frequencies. Having a voice codec like PilotEdge, where it sounds like you're talking on Skype or TeamSpeak, takes away from the experience in my opinion.

 

To each their own.

 

Take care,

 

Yes, to each their own... I hope there is something for both opinions in the future! For those who want to hear more clear and for those who want to hear more static and so on. At last I think the more people fly online, the more fun we all have.

[Fatih Yas 969587]

I heard a lot of real ATC , and there is nothing where we can match it with our codec. The real one is clearly to understand , and in vatsim many times its very bad. I fly also the region of myanmar , afganistan yes there is very bad but the rest at europe is not like in vatsim

[Andrew Morkunas]

New poll: Which will be released first?

 

• *Swift
  *New voice codec

Edited November 15, 2017 at 02:23 PM by Guest

[Magnus Meese]

128bit sim will come before either.

[Mike Lehkamp]

Optimistic are we?

[Robert Cardone 1323915]

I heard a lot of real ATC , and there is nothing where we can match it with our codec. The real one is clearly to understand , and in vatsim many times its very bad. I fly also the region of myanmar , afganistan yes there is very bad but the rest at europe is not like in vatsim

 

Most of the really bad audio on Vatsim is due to pilots and controllers using cheap microphones, not positioning the mic correctly, and not seeing how horrible they sound before signing on Vatsim. Some pilots sound they are using a mic that is in another room or facing away from them. Laptop microphones are the worst for sound. Using a decent headset mic is the first step to understandable audio Here is a good video explaining how important it is to set up your headset mic correctly. I also think that one someone has really bad audio, others especially the controller should bring that to their attention. When someone has to repeat a communication several times, it is obvious that their audio is not understandable.

[Andrew Morkunas]

Most of the really bad audio on Vatsim is due to pilots and controllers using cheap microphones, not positioning the mic correctly, and not seeing how horrible they sound before signing on Vatsim.

 

Testify brother, testify, say AMEN!!!!!

 

Granted the current voice CODEC is not the best but real world VHF Amplitude Modulation (AM) is not crystal clear either. Having flown on 'the other European network' using TeamSpeak2, I found the audio to be unrealistic because it was too good. The majority of the problems I have as a controller on VATSIM is bad audio from the pilots for the reasons mentioned above. Invest in good audio equipment (that means over $100) and you will sound mahvelous...

Edited November 20, 2017 at 08:39 PM by Guest

[Magnus Meese]

In my experience, the cheap voIP mics are usually better than expensive gamer headsets, as the high bitrate compressed to the degree VATSIM transmits can be disasterous. As long as it has a windshield and the placement is good, cheap mics can be great (so get a mic that's good enough, and STOP USING TEXT IF YOU DO NOT HAVE AN IMPAIRMENT!)

[Halldor Bui Jonsson]

this will bring me back to vatsim, so I cant hold back

I can't hold back either: You're not participating on the network just because of the codec..?

 

It´s not necceserily the quality of the transmission but the inherent lag in transmissions that´s hard to get over.

 

Having worked R/L radios for some time makes it doubly unbearable. Working tower, approach or a busy center with what is comparable in delay to working HF traffic under somewhat poor conditions takes the enjoyment out.

 

When this last relic of the Roger Wilco days ( some of you might not have been born when that was in effect) falls, so might my interest return in pushing virtual tin on VATSIM´s scope.

 

Hears to VATSIM finally pushing this. You might be surprised to know how many people who rely on voice comms have given up due to codec problems like this.

 

.. Back to my cave.

[Simon Kelsey]

It´s not necceserily the quality of the transmission but the inherent lag in transmissions that´s hard to get over.

 

Unfortunately whilst I don't want to sound miserable I should warn this is exactly the thing which cannot be fixed. Slightly improved, perhaps, but 'inherent' is a very good description of latency in VOIP systems involving multiple users spread all over the world on varying quality Internet connections.

 

At the end of the day, if you want to improve the quality of the audio and reduce drop-outs then you need more error checking. This allows the codec to reconstruct the audio from a smaller number of packets but by definition increases latency.

 

The other issue is that in a typical voice room you can have pilots from all over the world, connected to a server which could be anywhere in relation to them. So even ignoring the encode/decode time - it still takes me 500ms to ping Sydney and 30ms to ping London. Ergo if you send the same data to everybody at once you could have the best codec in the world but everyone will still receive it at slightly different times. Add in someone in a remote location with a flaky satellite-based Internet connection and they'll be looking at seconds worth of latency whatever the codec.

 

There will always be latency and there will always be the issue of people stepping on each other because everyone's received the message at slightly different times. Any new codec might be slightly more efficient but it will still suffer from the inherent issues of geography that are the main cause of voice latency now.

[Ross Carlson]

There will always be latency and there will always be the issue of people stepping on each other because everyone's received the message at slightly different times. Any new codec might be slightly more efficient but it will still suffer from the inherent issues of geography that are the main cause of voice latency now.

 

I don't think it's correct to say that geography (network latency) is the main cause of voice latency now. You talked about two of the sources of latency in typical VoIP systems, those being the encoding/decoding time in the codec and network transport latency. However, there is a third source of latency which is the buffering done on both the sending and receiving side in order to prevent gaps in the audio stream that can be caused by jitter in the network latency.

 

From what I've learned by talking with the devs that are looking into improving our voice system, it is this third source of latency that is the largest contributor for VATSIM. You said that latency is a problem that cannot be fixed, only slightly improved, but in actuality it can be dramatically improved. The amount of buffering done on both the sending and receiving end was designed to accommodate decades-old networking technology where there was far more jitter in the network latency. This buffering can be reduced now that connections with less jitter in the ping times are the norm. Yes, we will still have some users with high latency in their network connection, and they will still be prone to stepping on others or being stepped on, but those users are the exception rather than the norm these days.

 

The fact that different users begin hearing a transmission at slightly different times is not a problem. It's when that delay is relatively large for the majority of users that people end up stepping on each other frequently. With VATSIM's high amount of buffering, we are essentially guaranteeing that the delay will be large for everyone, not just people on high-latency connections.

[Halldor Bui Jonsson]

I think one important aspect to get right is to implement half-duplex transmissions. e.g. ATC wont hear incoming transmissions on the frequency he is transmitting on.

 

Same should be said with the transmitting pilot. If the two happend to transmit at the same time - do some sort of frequency blocking effect to any 3rd partie(s) that might be listening. The two would obviously be unaware though.

 

At least that would properly simulate frequency congestion rather than just a bunch of voices at the same time.

 

Such radio simulation already exists in some sims ( Falcon 4 BMS to name one). Most use some form of TS3 implementation.

 

On the delay aspect - It is just unbearable to me to have the xmit button light up 500ms to 1sec after initiating PTT. Even if the delay is in those ranges, when I press PTT, I expect to be able to start talking pretty much immediately. In other words it's not necessarily the lag itself but the delay from when I as a user want to do something until the time the system is ready to accommodate the request.

 

I would be so happy if the audio capture would be quick enough to properly send out a zipper. Whether the pilot hears it 500ms or 1 sec later is not really the key issue.

Edited November 22, 2017 at 11:23 PM by Guest

[Ross Carlson]

On the delay aspect - It is just unbearable to me to have the xmit button light up 500ms to 1sec after initiating PTT. Even if the delay is in those ranges, when I press PTT, I expect to be able to start talking pretty much immediately. In other words it's not necessarily the lag itself but the delay from when I as a user want to do something until the time the system is ready to accommodate the request.

 

I have found that your voice begins to be captured immediately after pressing PTT, and the delay in the TX light illuminating is superficial. Perhaps your experience is different? I would try a proper test to be sure. Press the PTT button immediately before rapidly counting "one two three four five" and see if others hear the word "one" fully.

 

I would be so happy if the audio capture would be quick enough to properly send out a zipper. Whether the pilot hears it 500ms or 1 sec later is not really the key issue.

 

On the contrary, the 500 or 1000 ms delay is at the heart of the issue. If that delay were shortened, we would have far fewer issues with pilots stepping on each other. The larger that delay, the greater the chance that two pilots will both discern that the frequency is idle, and start transmitting, not knowing that another user has already started transmitting. Obviously there will always be the possibility of one pilot transmitting before he begins to hear the audio from another user that is already transmitting, but the more we can reduce the delay between a user transmitting and other users beginning to hear it, the less we will have users stepping on each other. 500 to 1000 ms is WAY too much and creates far too much opportunity for stepping on other users.