Find audio sample rate for an HDMI signal or stream

[Themurf47]

I have a customer who is using an ATEM switcher, SDI output, going to a Lumantek SDI to HDMI converter and then to an Epiphan Webcaster 2 and on to the internet. The issue is that the Webcaster is throwing up the following message: Actual audio sample rate 44906.31Hz mismatched with expected (44100Hz or 48000Hz) Check your HDMI In audio settings.

When listening to the audio on a separate receiving device the audio is just clicks which is what I would expect with a mismatch of sample rate. If I use a different SDI to HDMI converter the error goes away. By using a different method, device or software I want to be able to see exactly what the Lumantek is actually pushing to the Epiphan device. Does anyone here know of a way that i can get an actual reading of the sample rate of the output of the Lumantek device.

[MJKlein]

Actual audio sample rate 44906.31Hz mismatched with expected (44100Hz or 48000Hz) Check your HDMI In audio settings.

Does anyone here know of a way that i can get an actual reading of the sample rate of the output of the Lumantek device.

I think that you already got that actual reading because the Webcaster error message gave you a very specific sample rate of 44906.31Hz. You got the exact sample rate in the error message. That is very cool. Looks like the Webcaster can be used to test devices but apparently I will only give you the sample rate if it's not 44.1kH or 48kH.

Obviously something is wrong with the clock on that device. But you can get the stream going by swapping it with a good converter. Other than a testing device, I have no idea how to measure the actual sample rate out of an audio device.

Welcome to the Livestreaming Technology Forums!

[Themurf47]

MJ, Thanks for the quick response. So to drop a little more information into the scenario, the manufacturer Lumantek has "tested" the actual unit used in the example above and have determined that there is no problem with their product and are refusing to support me by taking the unit back. I spoke at length with Epiphan regarding their unit and they also have no idea what the problem might be except to state what you stated, that the converter is modifying the sample rate.

The reason I am looking for an independent way to determine the actual sample rate is to be able to do my own independent test and report back to either Lumantek or Epiphan.

[MJKlein]

MJ, Thanks for the quick response. So to drop a little more information into the scenario, the manufacturer Lumantek has "tested" the actual unit used in the example above and have determined that there is no problem with their product and are refusing to support me by taking the unit back. I spoke at length with Epiphan regarding their unit and they also have no idea what the problem might be except to state what you stated, that the converter is modifying the sample rate.

The reason I am looking for an independent way to determine the actual sample rate is to be able to do my own independent test and report back to either Lumantek or Epiphan.

Holy cow, that's a shame. You sent it back to Lumantek and they tested it?

The main issue that I see with testing is that the vast majority of testers are cable continuity testers. You need something that will qualify a convertor and specifically the audio sampling rate. That is going to cost several orders above the price of the Unit Undergoing Test. There are also software validation solutions but this one, for example, is over USD $4,000:

https://www.keysight.com/us/en/product/ ... tware.html

This HDMI source controller is free. It might be worth seeing if it detects an issue with the audio sampling rate:

https://en.freedownloadmanager.org/Wind ... -FREE.html

This PDF by Tektronics is pretty interesting reading on what their software can test:

https://assets-us-01.kc-usercontent.com ... et-0_1.pdf

Ultimately your best advocate might be Epiphan. They have an interest in proving their gear is not the issue. I would talk to them about allowing you to send the converter to them to analyze. There may be some kind of firmware bug or other unforeseen issue that Epiphan could benefit knowing about and fix.

[Themurf47]

What I have discovered so far is that the neither the Lumantek nor the Epiphan units are at fault in this scenario. It appears, as reported by a company with experience in this arena, is that the Black Magic Designs ATEM switcher is outputting a non-standard reporting scheme that their SDI to HDMI converter can decipher but other manufacturer's converters can not. The Black Magic device will report through HDMI the correct sample rate while other manufacturer's devices can not read the data as supplied by Black Magic in the SDI file and therefore are reporting an erroneous sample rate. This, then appears to be a case of Black Magic not playing with the same rules as everyone else. I guess that helps sales.

[MJKlein]

Wow! What a find! Yeah, that would do it. Have you notified Black Magic? They take a lot of pride in being a high end company. I'm sure they would be very interested in this issue. But then again, maybe they are trying to encourage users to deploy Black Magic gear exclusively.

I apologize for taking so long to approve this message. Each new user has a 5 message threshold that requires approval (to suppress SPAM of course). This thread is very useful and will benefit others in the future. Thanks so much for posting.