First of all, please download and install SPEK: http://spek.cc/ , the free open source code Spectrum Analyser.
Then please let me teach you how to interpret SPEK spectra here below:
1. Please look at picture 1. Here is a webm stream downloaded from a YouTube video. Most webm streams contain a Vorbis Ogg soundtrack, but some webm streams including this one contains an Iibopus Opus soundtrack instead. (And both types are way better than AAC soundtracks uploaded after March 2016).
Can you see that the upper right red arrow is at the level 20 kHz, if you please take a look at the left vertical scale? Why is it at 20 kHz? Well, this is because this person finally saved or dubbed his video at 20 kHz sound quality.
Can you see that lower left red arrow is at the level 18 kHz? This is because this person a bit earlier saved his off-vocal instrumental soundtrack at the sound quality of 18 kHz. This means that the actual music in this download has plenty of music up to 18 kHz, but no music above 18 kHz. Young human ears stop at 20 kHz. Dogs can hear much higher sounds and music than 20 kHz,
2. Please look at picture 2. Here is a mp4 stream downloaded from the same YouTube video. Virtually all mp4 streams contain an AAC soundtrack, including this one. YouTube slashed the sound quality of new HD uploads from mp4 AAC 192 kbit/s HiFi to 126 kbit/s LoFi on March 28th 2016.
Can you see that the red arrow is at the level 16 kHz? This means that this video was uploaded after March 28th 2016. The red arrow used to be at 18 kHz before that date.
Now let's think for a bit. Suppose that you have downloaded both the video stream in picture 1 and 2. Which one should you use? #1 has a higher bitrate, 160 kbit/s versus merely 126 kbit/s in #2. And the music at 16-18 kHz is missing in #2 but present in #1. So you should obviously use #1. (Or rather, you should download both, examine both with SPEK, and then select the one that happens to look better in SPEK, which happens to be #1).
3. Please look at picture 3. Here is an à capella soundtrack made by a Russian girl using her own voice.
Can you see the upper left red arrow at the 20 kHz level? This is because she recorded her own voice singing the main vocals at 20 kHz sound quality.
Can you see the lower right red arrow at 16 kHz? This is because she recorded her own mumbling level background vocals with 16 kHz sound quality.
4. Please look at picture 4. Here is her completed mixing of her own mp3 soundtrack intended for her soon to be YouTube upload. (What you can see in picture 4 is not something that was downloaded from YouTube by me. Instead it's a mp3 file that she placed at a file sharing site.)
Can you see the upper right arrow at the 20 kHz level? This is because her own vocals were recorded with 20 kHz sound quality.
Can you see the lower left arrow at 16 kHz? Oh damn it, so sad, this is because she used the mp4 download in picture 2, instead of the webm download in picture 1. Her mistake was that she didn't use SPEK to examine every imaginable possibility during her video production.
Everyone, please remember this simple RULE: Always use SPEK to examine your soundtrack at EVERY step of your sound and video production, to avoid unnoticed losses of sound quality.
Please note that the red arrows were photoshopped in place by me on June 7th 2020. So red arrows appearing in interesting places are not a feature included in SPEK. Thus, this sentence:
"The red arrow used to be at 18 kHz before that date."
"The red arrow used to be (needed to be photoshopped in) at 18 kHz before that date."
And here is a guide to the actual soundtracks used in this slideshow tutorial:
1. https://www.youtube.com/watch?v=ZpWiQYRuhKE only if you click on "webm" in your video grabber
2. https://www.youtube.com/watch?v=ZpWiQYRuhKE only if you click on "mp4" in your video grabber
[Note that no. 1 & 2 is the very same video]