Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Strange frequency spike near 18kHz in song (Read 8279 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: Strange frequency spike near 18kHz in song

Reply #25 –
Anyway, I've managed to obtain a program called Sigview and I've got it to generate a 3D spectrogram.

Spoiler (click to show/hide)

As expected, there's a prominent spike in that higher frequency region for most of the song.
Lossless: flac --best --verify
Lossy: opusenc --bitrate 160

Re: Strange frequency spike near 18kHz in song

Reply #26 –
On my ABBA recordings (found in some of the normal distributions also) -- I think that SuperTrouper has it a lot of the time -- there is a 19.2kHz bit of fuzz across the spectogram.  I had guessed that SPECIFIC problem was some kind of serial interface noise leaking into the audio.  For some of my processing to avoid litting even the smallest amount of distortion to occur, I notched that out with no audible effects.  (I used an FIR style filter instead of a recursive one -- a bit better control as to the removal, and no phase issues.)

John

Re: Strange frequency spike near 18kHz in song

Reply #27 –
The initially posted spectrogram had better time resolution, and showed the hi-hat keeping a pattern, which aligns with the rest of the music and the bass notes at the bottom. In the 3D waterfall, the spike is misleadingly shown to be almost constant.

My interpretation is that the spike with its wide sidebands is the whole synthetic hi-hat, generated maybe by a younger listener, or maybe they never looked at the spectrum analysis and just tweaked it until the audible portion was as desired. The entire noise pulse is centered around its peak, and has no other features. The "hi-hat" could be made with a high Q bell filter on noise. I can hear the longer tsss sections at comfortable volume just fine, but not the two shorter hits.

The tones in Super Trouper are unlike this: narrowband and too quiet to be heard directly or as distortion products (they're even quieter than the video monitor tone on most recordings).

High frequencies are common across electronic, artificial music, including old. Jean-Michel Jarre "Equinoxe" is very bright in its original CD issues, and most of them distort during loudest moments.
 

Re: Strange frequency spike near 18kHz in song

Reply #28 –
Quote from: jsdyson on
On my ABBA recordings (found in some of the normal distributions also) -- I think that SuperTrouper has it a lot of the time -- there is a 19.2kHz bit of fuzz across the spectogram.  I had guessed that SPECIFIC problem was some kind of serial interface noise leaking into the audio.  For some of my processing to avoid litting even the smallest amount of distortion to occur, I notched that out with no audible effects.  (I used an FIR style filter instead of a recursive one -- a bit better control as to the removal, and no phase issues.)

John
Hm, that's interesting. I think I have an ABBA CD somewhere (their Gold hits album, the one with Waterloo as the last track, track 19), I'll have to see if it's on there. A question though ... What kind of processing were you doing to not let such a high frequency interfere with whatever you were doing?
Quote from: j7n on
The initially posted spectrogram had better time resolution, and showed the hi-hat keeping a pattern, which aligns with the rest of the music and the bass notes at the bottom. In the 3D waterfall, the spike is misleadingly shown to be almost constant.

My interpretation is that the spike with its wide sidebands is the whole synthetic hi-hat, generated maybe by a younger listener, or maybe they never looked at the spectrum analysis and just tweaked it until the audible portion was as desired. The entire noise pulse is centered around its peak, and has no other features. The "hi-hat" could be made with a high Q bell filter on noise. I can hear the longer tsss sections at comfortable volume just fine, but not the two shorter hits.

The tones in Super Trouper are unlike this: narrowband and too quiet to be heard directly or as distortion products (they're even quieter than the video monitor tone on most recordings).

High frequencies are common across electronic, artificial music, including old. Jean-Michel Jarre "Equinoxe" is very bright in its original CD issues, and most of them distort during loudest moments.
Yeah, sorry about that; I still need to figure out how to use this software to make it generate better graphs; as it is, since this is an older laptop, generating this image already took a while.

It's fairly clear this sound is somehow part of the percussion synth based on how it's intermittent and happens to stop completely when the percussion stops and the song fades out with held synth notes, so I agree with your analysis of it being a synth hi-hat.

As far as age is concerned, the person responsible for this track (and others on the same game OST) is a fellow by the name of Tee Lopes; when the question came up in an earlier post on this thread, I tried to google to see if his age is listed somewhere publicly, but alas, I can't seem to find it, so *shrug*
Lossless: flac --best --verify
Lossy: opusenc --bitrate 160