Decoding digital voice with hardware possible?

Discussions about everything to do with SDRuno
Post Reply
DL2DW
Posts: 2
Joined: Wed Dec 07, 2016 10:45 pm

Decoding digital voice with hardware possible?

Post by DL2DW » Tue Jan 03, 2017 9:32 am

There is the opportunity to hear digital language (DMR, D-star, etc.) with SDRuno and DSD. But the quality is not particularly great. Hence my question whether there is a possibility of hardware? For example a USB module with AMBE3000 decoder. He could then decode the language.

And sorry for my bad English, but the universal translator is guilty. :oops: :D
Last edited by DL2DW on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

mikael
Posts: 18
Joined: Tue Dec 06, 2016 9:03 am

Re: Decoding digital voice with hardware possible?

Post by mikael » Tue Jan 03, 2017 10:52 am

DSD uses the authors mbelib as speech codec. You can of course replace it with a hardware decoder, but I don't believe this is your problem. Instead it's more likely the levels of the audio stream from SDRuno not agreeing with the DSD input filters. First check the error rates DSD presents. If none, you've got the best sound quality achievable, and then can begin looking at the speech codecs performance. Else try tuning everything - it can be a hazel due to the lack of feedback from the deciding process. Also have a look at DSD+ as it is a more developed product than the open source DSD.

Reason: No reason

mikael
Posts: 18
Joined: Tue Dec 06, 2016 9:03 am

Re: Decoding digital voice with hardware possible?

Post by mikael » Tue Jan 03, 2017 11:09 am

Oh, one clarification; the speech codec is only the last step of decoding digital voice modes. This is not very likely the step that causes your problems, but the earlier steps of the fsk bit stream demodulation. The AMBE3000 only handles the speech decoding.

Reason: No reason

DL2DW
Posts: 2
Joined: Wed Dec 07, 2016 10:45 pm

Re: Decoding digital voice with hardware possible?

Post by DL2DW » Thu Jan 05, 2017 6:34 pm

I bought myself the AMBE3000 shield DV-mega now. And I've done the test with a wav file, recorded from my ICOM transceiver. The voice quality through the AMBE3000 chip is to lengths better than with DSD or DSD+.

And above all, the latency is MUCH lower. With the Ambe3000, which runs almost in real time, the DSD + hangs almost 2 seconds behind (on a core i7 QuadCore)

Now I gotta just yet somehow it with SDRuno together.
Last edited by DL2DW on Thu Jan 01, 1970 12:00 am, edited 0 times in total.
Reason: No reason

Post Reply