ProVoice sucessfully decoded with DSD 1.3! in Palm Bay FL

[mhzscan]

I have successfully decoded ProVoice using DSD 1.3!!! I got a 97% rate on the voice recovery, that is not bad at all considering I was using the MIC input instead of the preferred LINE input. I tested this on the Palm Bay Police Provoice Talkgroups and it worked great! Details & info are pretty hard to find on how DSD is configured to work. The wiki doesnt really state that much and the README is very technical, even for me and im a Microsoft professional. The forums however were a bit helpful.

Soooo, basically DSD runs on Linux and decodes the digital audio on a voice channel using a discriminator tap. Its NOT decoding the control channel only the voice portion of it. The tap is connected to a computer via a LINE or MIC input. DSD and an app called Mbelib do the rest. I believe this will only work on conventional scanners and not the trunking scanners, since trunktrackers will automatically ignore provoice transmissions. Again, my understanding is that tracking will have to be done manually for now.
Also XT's C_CH feature will NOT work with DSD.

Maybe in future versions, there will be a way to get it to work with trunktrackers.
DSD looks very promising and considering its open source, I wouldn't be too surprised if Uniden incorporates this to it's future TT lineup.

Wiki Link:
Digital Speech Decoder (software package) - The RadioReference Wiki

Helpfull Forum Threads:
http://forums.radioreference.com/tr...oding/179709-dsd-1-3-mbelib-1-2-released.html

http://forums.radioreference.com/tr...505-need-help-installing-dsd-into-ubuntu.html

http://forums.radioreference.com/tr...l-speech-decoder-1-2-mbelib-1-1-released.html

 

[k3ocm]

If only DSD could me made to work with the Trunker software

 

[AlexC]

Very cool.

The best way to do this is using a USRP2 and a WBX board... If you can capture the entire trunk system you can then decode everything... It's going to require some processing power but you could essentially listen to every single active channel on the system at one time...

Or, use two wbx boards (or any two rx boards in your choice of band...) Have the control channel sitting on one RX and write the code to steer the second RX unit based on the decoded control channel data. The first option is much cooler, but requires a lot of processor power and hard drive "scratch" space to pull it off.

-alex

 

[SRSP2282]

you could essentially listen to every single active channel on the system at one time...

Alex you just blew my mind! If only I had a ProVoice system near me to play around with...

 

[mtindor]

I believe this will only work on conventional scanners and not the trunking scanners, since trunktrackers will automatically ignore provoice transmissions. Again, my understanding is that tracking will have to be done manually for now.

Just to clarify, the only time a trunktracking scanner would not hear the digital voice to decode it is if it were in trunking mode and it knew to ignore Provoice talkgroups because it couldnt' decode them.

This does not mean that you can't manually tune the voice channel with a trunktracker and then pipe the raw voice channel audio into DSD.

So, regardless of whether you have a conventional or trunking, analog or digital scanner, as long as the scanner has the capability of tuning the actual frequency of the voice transmissions and as long as you have a discriminator tap, you can pipe the raw audio into DSD for processing.

Mike

 

[cja1987]

Forgive my ignorance, but the way I read this, you really can't decide what talkgroups you want to listen to, you can just decode whatever is being carried across a Pro Voice capable voice channel?

Do I understand correctly or is there a way to discriminate?

Not a problem in Brevard County because relatively few pro voice talkgroups exist but in a heavily pro voice area, it would get annoying not being able to choose talkgroups to listen to. I assume this is what you guys are talking about when you mention possibly getting it to work with trunker software?

Hopefully this does not push agencies to add ESK

 

[bcradio]

ESK wouldn't do anything to the situation you describe... If you want to follow a particular talkgroup, I think you'd need two computers and two discriminator tapped scanners (one of which would need to be computer programable and operatble)

One dis. tapped scanner to run a program like Etrunker or Unitrunker... This will control the the second disc.tapped scanner to jump to individual voice channels on which DSD would give you the audio.

 

[mhzscan]

So, regardless of whether you have a conventional or trunking, analog or digital scanner, as long as the scanner has the capability of tuning the actual frequency of the voice transmissions and as long as you have a discriminator tap, you can pipe the raw audio into DSD for processing.

Mike

Thanks for clarifying that, I was meaning to say.

In regards to getting it to automatically track the TGs, I'd be interested in trying the two scanner/two tap scenario scenario. Unfortunately I only have on scanner that is tapped. Either way, someone is bound to develop a app that will make this work similar to trunker/etrunker. Time will tell.

In regards to AEGIS, I think someone has already submitted audio samples of AEGIS to the DSD author... Will see. (oops this was meant to be response for another thread.)

 

[mtindor]

Thanks for clarifying that, I was meaning to say.

In regards to getting it to automatically track the TGs, I'd be interested in trying the two scanner/two tap scenario scenario. Unfortunately I only have on scanner that is tapped. Either way, someone is bound to develop a app that will make this work similar to trunker/etrunker. Time will tell.

I envy you guys with ProVoice systems nearby you... You get a chance to do something most people will never have a chance to do, decode Provoice. Of course, you also now have the benefit of listening things that some agencies on the system probably thought you'd never get a chance to listen in on

Mike

 

[Thayne]

I just had a thought that I am fairly sure that the old Pro-92 would stop on ProVoice transmissions, but if course it did not have a Disc Tap or anything. I had 2 of them but gave them away years ago.

If anyone out there still has one like in Florida or somewhere else that uses ProVoice but not ESK it would be easy to verify if my memory is correct. It would be very easy to put a Disc tap on one.

Being that the "cat is out of the bag" I don't see any reason why we won't see a scanner that will do it sometime in the next few years. As long as the method of doing it does not impinge on the DVSI patent(s), I think scanner manufacturers would be in the clear. I can take all the naysayers saying I am full of it, so lets hear what anybody thinks---
I don't care myself, because I already have a portable & 2 mobiles with PV working.

 

[saber2k5]

The Pro 92 does indeed stop on PV and even displays the TGID.

 

[i5adam8]

I just had a thought that I am fairly sure that the old Pro-92 would stop on ProVoice transmissions, but if course it did not have a Disc Tap or anything. I had 2 of them but gave them away years ago.

If anyone out there still has one like in Florida or somewhere else that uses ProVoice but not ESK it would be easy to verify if my memory is correct. It would be very easy to put a Disc tap on one.

Being that the "cat is out of the bag" I don't see any reason why we won't see a scanner that will do it sometime in the next few years. As long as the method of doing it does not impinge on the DVSI patent(s), I think scanner manufacturers would be in the clear. I can take all the naysayers saying I am full of it, so lets hear what anybody thinks---
I don't care myself, because I already have a portable & 2 mobiles with PV working.

I totally agree with you. I really believe that it wont be very long until we see a scanner that can track provoice.Now the next challenge will be Opensky lol

 

[mhzscan]

The Pro 92 does indeed stop on PV and even displays the TGID.

Really? Time to look one up on ebay. Thanks for the heads up.

 

[mhzscan]

For those with a ProVoice system close by here in Florida, another member in RR has come up with a solution on how to get unitrunker to work with DSD. Below is his post:

Unitrunker wrote

"I've gotten a lot of requests to add DSD decoding capabilities to Unitrunker. In the current US Patent Law climate - that won't happen. Sorry folks.

But ... I have been also hearing from others who successfully use DSD and Unitrunker to follow voice calls - most notably on EDACS ProVoice systems. Here's how:

1. Windows machine running Unitrunker configured for (at least) two receivers
a. connected to a signal receiver monitoring the control channel (may be disc. tap or one of the "Inline" models).
b. connected to a control receiver configured in the Voice role; this radio must have a disc. tap.
2. Linux machine running DSD
a. DSD decodes the discriminator audio from receiver in 1(b) above.

I know - this requires two machines. However, it is possible to run one or the other as a VM host using VMWare or whatever virtual machine environment you like. WINE might be an option for an all-Linux solution. I haven't tried that."

 

[OldDeadOne]

Interesting thread,I've subscribed to this(just for future reference if ever). Would love to see a video how it's done.

 

[scg]

Will all PRO-92's do this or only the original version? I thought I read where the A and B versions did not work for this.

 

[dic]

mhzscan

Thank you for your service to the scanning community.
I have to wonder about two things:

1-Are all the innumerable posts saying that NO ONE WILL EVER BE ABLE TO DECODE PROVOICE!!!! going to be removed from this forum? (That is a JOKE by the way)

2-Am I the only one who is COMPLETELY LOST trying to figure out how this has actually been achieved? I am a pretty intelligent General Class Ham, Commercial Pilot and pretty good with computers, and I can't even begin to understand what is happening.

Maybe it's me.

Thanks again,

Dic
N1XBA

 

[mhzscan]

Dic thanks for the kind words, however I need to make it clear that I'm NOT the person who created or came up with this new development. I'm simply testing the software that was developed by someone else and sharing my findings. The anonymous developer goes by the name of DSDAuthor, you can find his postings in the Trunking Control Channel Decoding Forum.

The easiest way to explain how DSD works is that the DSD software will listen to the RAW digital noise that is spewed out on the voice channels, NOT the control channel, via a discriminator tap. Then DSD will process/decode this digital speech format and send it to the mbelib vocoder application which will then synthesize the decoded digital speech.

More detail info can be found in the DSD wiki below:
Digital Speech Decoder (software package) - The RadioReference Wiki

This is a work in progress, originally it started with a couple of digital formats. The last revision included Provoice. Now there is talk of including formats like OpenSky (4slot format) and possibly TETRA.

Hope this helps.

MHZScan

 

[SCPD]

I'm just a scanner guy and know little of all this techno stuff but I have to wonder if there is an issue of the legality of monitoring public safety communications intended to be encrypted? Not that I wouldn't jump on board but I just would like to know what the risks are? It's one of those cases where do I wanna be FREE and wish I could hear an agency or be in JAIL wishing I could be free?

 

[mhzscan]

No need to worry, Provoice is NOT encryption. However some agencies use provoice and then encryption, like the state of Florida.

MHZscan