Using Audacity to save songs I compose. Playback on YouTube almost too faint to hear!

[nelsonkane]

There could be a million reasons for this. In your opinion, what is most likely? Any suggestions appreciated! Do most people have such a difficult time making YouTube animated musical cartoons?

1. If most of my video content is hundreds of hand-created pictures using Paint or Paint 3D, could the Paint programs be decomposing and have a negative affect on the audio?

2. If I’m not recording each individual song at enough volume, is that the problem? In any rock music song, there will be some parts that are 95-100% volume, and many parts that are 50% of the allowable volume. Do I need to record the songs on Audacity so that every note is equal volume?

3. Should I not be using Clipchamp at all? Should I not be using Photos Legacy at all?

4. I’ve never used “compression.” Is that of any value?

5. Suppose I recorded every note on Audacity, at an equally-high
   volume? Would it all just sound like "fuzz?

[DVDdoug]

The video shouldn’t affect the audio.

Hang-on, “loudness” gets complicated…

Note that YouTube and all of the popular streaming services use loudness matching (1) so the loudest part of your audio should (approximately) match the loudest part of everybody else. This is a linear adjustment before playback starts so the quiet parts remain relatively quiet and loud parts remain relatively loud.

Make sure you Normalize or Amplify to “maximize” the digital level. (The “digital maximum” is 0dB.) These are linear adjustments (like adjusting the volume control) so as long as you don’t “try” to go over 0dB it won’t affect sound quality. This should generally be done as the last step.

But, peaks don’t correlate well with loudness, so for example, if you peak-normalize all of your music (2) some songs will still be louder than others. Short-term peaks don’t sound as loud as sustained loudness and our ears are less sensitive to low & high frequencies, and most sensitive to high frequencies.

Certain instruments (acoustic guitar for example) are highly dynamic. There is a wide range between loud & quiet and if you normalize/maximize it, it will sound quiet compared to most music. Saturated-distorted electric guitar is more “dense” and will sound loud when normalized.

Very-deep bass or strong highs can also limit the “loudness”. The highs usually aren’t an issue unless they are artificial/synthesized.

YouTube (and the others) won’t push the peaks into clipping (distortion) and that’s probably why you are quieter than everybody else.

I’ve never used “compression.” Is that of any value?

YES! Dynamic compression is how you “win” The Loudness War! (Although you can’t really win with online loudness matching.)

In general, compression lowers the loud parts and/or boosts the quiet parts, pushing everything toward the same volume.

In practice, it usually pushes-down the loud parts and then make-up gain is used bring-up the overall volume.

Limiting is a kind of fast-compression. For “loudness”, limiting usually works better than regular compression because there is no delay (no “attack time”).

Audacity’s limiter actually uses look-ahead (negative attack time) so it’s very-good and it doesn’t distort the wave shape. It also has a make-up gain option.

Compression (including limiting) does, of course, reduce the musical dynamic contrast and if over-done it can make music boring. A lot of people really hate the loudness war for that reason!

Virtually ALL commercial music has SOME compression except for maybe some classical recordings.

Don’t confuse dynamic compression with file compression like MP3. MP3 has no effect on the dynamics.

Suppose I recorded every note on Audacity, at an equally-high volume

Again, music is pretty boring with no dynamics.

Would it all just sound like "fuzz?

Only if it’s “too loud” and clipping/distorting. And if there are multiple instruments or chords it will be louder whenever there is more than one “note” combined.

(1) Sometimes this is called loudness normalization, but “normalization” usually means peak-normalization which is a “simple mathematical” process with no regard to perceived loudness.

(2) Most commercial music is already peak normalized but some songs are still louder than others.