You are here

How can I detect audio dropouts without listening to a sinewave for several hours?

For anything relating to music-making on Windows computers, with lots of FAQs. Moderated by Martin Walker.

Moderator: Moderators

How can I detect audio dropouts without listening to a sinewave for several hours?

Postby shufflebeat » Wed Apr 28, 2021 5:26 pm

I'm testing a couple of recording drives on the Qu Pac and would like to check for dropouts. Logic suggests first recording a sinewave and looking back at the waveform would be straightforward but the Qu doesn't generate sine at inputs, also life is sometimes unexpectedly short so don't want to spend it listening to 500Hz for hours, however perfect it is.

Any thoughts?

I'm in Window Music because any testing setup will be done in W10 (or maybe iPad).
shufflebeat
Jedi Poster
Posts: 6213
Joined: Sun Dec 09, 2007 1:00 am
Location: Manchester, UK
"Dance, dance. wherever you may be, for I am the Lord of the damp settee..."

Do yourself a favour, wear earplugs at gigs.

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby Albatross » Wed Apr 28, 2021 7:05 pm

How about a copy of Audacity and then use the 'search silence' ... you can set the parameters and it will find - the sound - of silence.
Albatross
Frequent Poster
Posts: 1590
Joined: Sun Jul 26, 2015 12:00 am
Location: Airstrip 1, Eurasia

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby shufflebeat » Wed Apr 28, 2021 7:24 pm

I like that but I suspect a dropout in this instance wouldn't take the form of actual silence.
shufflebeat
Jedi Poster
Posts: 6213
Joined: Sun Dec 09, 2007 1:00 am
Location: Manchester, UK
"Dance, dance. wherever you may be, for I am the Lord of the damp settee..."

Do yourself a favour, wear earplugs at gigs.

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby RichardT » Wed Apr 28, 2021 7:37 pm

Why don’t you set it recording something for an hour, then do it again, then null the two files and see if you have any signal.

I’m not sure what a Qu PAC is, so ignore me if that makes no sense...
RichardT
Frequent Poster
Posts: 851
Joined: Fri Aug 13, 2004 12:00 am
Location: London UK

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby blinddrew » Wed Apr 28, 2021 7:49 pm

Would the kind of drop out we're talking about here trigger a signal spike? If so you could set your sine wave at a low-but-consistently-measurable level (I dunno, -30dB maybe) and leave it playing. If you come back to levels that have a peak much higher up then you know something has glitched?
User avatar
blinddrew
Jedi Poster
Posts: 14525
Joined: Sun Jul 05, 2015 12:00 am
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby James Perrett » Wed Apr 28, 2021 9:35 pm

I'd do something like record a sinewave and then do a frequency analysis of the recorded waveform (or just look at it on a spectral display). Any glitches will immediately show up as spikes.

This can be easily done using the JS Tone Generator in Reaper to generate a tone and Reaper's Spectral View to display the result. Other software probably has similar facilities - certainly Adobe Audition can do the same thing.
User avatar
James Perrett
Moderator
Posts: 10644
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby merlyn » Thu Apr 29, 2021 1:23 pm

James Perrett wrote:I'd do something like record a sinewave and then do a frequency analysis of the recorded waveform (or just look at it on a spectral display). Any glitches will immediately show up as spikes.

The fourier transform of a sinewave is a spike. An impulse (which a dropout is an approximation of and does look like a spike in the time domain) looks like a horizontal line in the frequency domain. An ideal impulse (a Dirac delta) contains all frequencies at an equal amplitude.

Given that a practical digital audio system is band-limited a dropout looks like horizontal fuzz across the whole spectrum in the frequency domain.
merlyn
Regular
Posts: 280
Joined: Thu Nov 07, 2019 3:15 am

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby James Perrett » Thu Apr 29, 2021 2:16 pm

I wasn't clear - on a spectral display in a DAW any glitches show as vertical spikes while the continuous tone shows as a horizontal line. This is opposite to a spectrum analyser view where the tone is a vertical spike.
User avatar
James Perrett
Moderator
Posts: 10644
Joined: Mon Sep 10, 2001 12:00 am
Location: The wilds of Hampshire
JRP Music - Audio Mastering and Restoration. JRP Music Facebook Page

Re: How can I detect audio dropouts without listening to a sinewave for several hours?

Postby merlyn » Thu Apr 29, 2021 3:02 pm

I think it's worth remembering that a spectral display has the fourier transform on the y-axis and time on the x-axis with colour being the third dimension of amplitude. Just to keep it all straight in one's head. :thumbup:
merlyn
Regular
Posts: 280
Joined: Thu Nov 07, 2019 3:15 am