as a follow-up, here's a video and a podcast for Sevana's product presentation: http://www.voipusersconference.org/2015/vuc567-sevana-evaluating-call-quality/
On Fri, Nov 6, 2015 at 11:44 AM, Stanislav Sinyagin <[email protected]> wrote: > Greetings, > > Here's a short follow-up to yesterday's presentation of Zebbra guys. > > What Markus and Roland have presented, is an approach to monitoring > the quality inside the box (CUCM in their case). But there are > actually ways to monitor the quality outside of the box, and they have > been around for a while. > > For example, I've seen a demonstration of Malden MultiDSLA tool over > 10 years ago. At that time they only had the analog interface that > emulated a handset, and two Malden boxes were sending an audio sample > to each other and assessing the quality of received audio. The tool is > quite expensive though. > > There are currently several commercial solutions for audio quality > analysis: > > POLQA (expensive), recommended by ITU-T: it compares two audio files > and produces the quality assessment score. > > Sevana AQuA (less expensive): it also compares two audio files and > produces several quality metrics. I compared it with PESQ (predecessor > of POLQA), and AQuA works significantly faster. It was very helpful in > detecting lost RTP packets in the tests where I could not place a > packet sniffer wherever I wanted. > > There are also various approaches to passive quality analysis, and > some are documented in ITU-T P.563 and G.107. Sevana is also offering > its PVQA tool that tries to detect packet loss and audio distortions > by analyzing the input audio. > > I could not find any open-source tools which would help in quality > analysis at the audio level. But there's plain old tshark which > produces the loss and jitter statistics for RTP streams, and it's > pretty efficient if you are able to capture the traffic at the > receiving end. Also VoIP monitor provides the ability of G.107 > analysis on captured data packets. > > Here's my detailed article on using Sevana AQuA: > https://txlab.wordpress.com/2015/06/02/quality-assurance-for-voip-calls-2/ > > And here's my demo lab, and you have a possibility to send an audio > recording for Sevana PVQA analysis: > http://voxserv.ch/demolab.html > > cheers, > > -- > Stanislav Sinyagin > Senior Consultant, CCIE #5478 > [email protected] > +41 79 407 0224 > -- Stanislav Sinyagin Senior Consultant, CCIE #5478 [email protected] +41 79 407 0224
_______________________________________________ swinog mailing list [email protected] http://lists.swinog.ch/cgi-bin/mailman/listinfo/swinog
