I have been asked to make several three and one half hour playlists of music from the 50’s through 1980. Each song comes from a different source, and thus the volume levels vary. I made the playlists in iTunes (their Sound Check feature, which is supposed to correct this, is basically useless) and imported them into Audacity. I am very new to this program and have some very basic questions. On the first playlist there are 70 selections, carefully sequenced. I cannot see how i play one after the other. Is that possible? Initially I found that ‘the terrible noise’ I heard was because all 70 tracks were playing simultaneously!
Obviously what I want to do is to maximize the level on some tracks to match the louder ones. I understand that I cannot do a ‘blanket’ adjustment, but have to treat each tune separately. How do I do that? I normalized all the cuts as I imported them, thinking that that would be a start.
Sound Check has two parts. First the files must be scanned and tagged by Sound Check. then on playback, if the media player supports Sound Check, then the player adjusts the playback volume automatically. Both parts are required for Sound Check to work.
Also, assuming that Sound Check has similar features to Replay Gain, there will probably be an “album mode” and a “track mode”. For what you want, you need to use “track mode”.
Yes that’s possible.
Press Command + A to select all
“Tracks menu > Align Tracks > Align End to End”
However, for the purpose of balancing the tracks, it may be easier to just use the track “Solo” button to listen to one track at a time. This allows you to skip from one track to another while playing. See: Audio Tracks - Audacity Manual
No, the other way round. If you try doing that then it is very likely that you will push some of the quieter tracks into distortion.
Then find the quietest track. Because you have brought all of the tracks up as loud as they will go without distorting, this track cannot go any louder. So now you need to reduce the level of other tracks to match this level. To adjust the level, use the track Gain slider (Audio Tracks - Audacity Manual)
It’s open source, free, available for Windows, Linux and Mac and has a lot of built-n features that are meant just for your application. And it uses the iTunes library. I don’t know if it uses the volume correction, but it’s worth trying it out.
Cyrano,
Thanks for the Mixx tip. I have downloaded it and taken a quick look, and it does seem as if it may do what I want it to do. However, as I have so much time already invested in Audacity, I am going to follow Steve’s recommendations today and see where I get.
Bob
Then find the quietest track. Because you have brought all of the tracks up as loud as they will go without distorting, this track cannot go any louder. So now you need to reduce the level of other tracks to match this level. To adjust the level, use the track Gain slider (> http://manual.audacityteam.org/man/audi > … .html#gain)
All tracks have been normalized.
I tried to find the quietest track by going to View>Mixer Board and selecting all the tracks and playing them simultaneously. With the volume off - sounded nicer that way. Then I looked at the peaks registered on the meters and found one track lightly quieter. Yet when playing it solo, the output at the top of the Audacity window showed it to be just as loud as anything else. Is there a better way to determine relative volumes? Doing over 70 tracks by ear is not feasible.
Looks like you have “scrolled” over the track’s vertical ruler. It can happen quite easily by accident if you have a mouse with a scroll wheel (or equivalent).
See here for the scroll / zoom gestures: Audio Tracks - Audacity Manual
The quickest way back to “normal” is “Shift+Right Click” over the vertical ruler.
If you normalized the tracks, then they will all have the same maximum peak level - that’s what normalizing does.
Peak level is NOT the same as “loudness”. Loudness is subjective, not an absolute measure. The only really accurate way to determine whether one track is louder than another is to listen to them.
Listening is the best way. Because loudness is a matter of how we “perceive” the sound, it is the only really “accurate” way - and your idea of “loud” may be slightly different to my idea of “loud” (though generally people tend to agree more or less, so long as they do not suffer from hearing problems.
Well there is an “approximate” way that may be close enough. I wrote a plug-in that attempts to adjust the levels of tracks to about the same “loudness”. In my tests it is usually fairly close, but you can let me know what you think. The plug-in is available here: ReplayGain plug-in
The download to use is the link that says “New Version”.
The only documentation for that plug-in is in that forum topic.
Peak level is NOT the same as “loudness”. Loudness is subjective, not an absolute measure. The only really accurate way to determine whether one track is louder than another is to listen to them.
Obviously, you are quite right. I was trying to determine the quietest one out of seventy. That is what I meant by it not being feasible to remember, for example, the volume of track 12 versus track 39. Guess I start at the beginning and go through them sequentially. I’ll check out your plug in - thanks for the tip.
and your idea of “loud” may be slightly different to my idea of “loud” (though generally people tend to agree more or less, so long as they do not suffer from hearing problems.
I’ve lost close to half my hearing, but loud is comparative! As long as all are at the relatively same volume, all will be well. The big problem here, of course, is that these tracks have been mastered by 70 different folk, so they are by definition going to sound ‘different’. But that is GOOD!
On two or thee occasions when playing a tune, it will just stop. Without touching the mouse, totally hands free. I go back to the song, and start it again, and it plays all the way through perfectly. This has happened on different songs, not the same one.
If there is a “selection” (part of the track highlighted) then play will begin at the start of the selection and stop at the end of the selection.
Nothing selected or highlighted that I am able to see. I am playing heads and tails to compare volume … sometimes I let the entire track play through. Usually I listen to the first twenty / thirty seconds, and then to the last ten / twenty seconds, and then into the next one … it seems to be when I let the track play through when it happens. As long as it isn’t actually in the file, I guess it doesn’t matter.
Have you got all 70 songs imported at once? That might be asking rather a lot.
If you only have one track then what playback device are you using? “Audio to buffer” in Audacity’s Recording Preferences might be set too low - that setting affects playback too.
I mistakenly clicked on Time Shift and now cannot ‘unclick’ it. How do I do that? I looked in the manual, but it doesn’t seem to address that issue.
Thanks,
Bob
I am now at the Export stage. I chose Export Multiple from the File menu, and designated a Folder & location. After clicking Export, the Metadata box comes up, which gives me a chance to tidy up the info on each file. As expected, Track 1 came up first, and then track 3. As all my tracks are numbered sequentially, I cannot understand what happened to it. It is still present in the playlist.
I want to move on with this project, but am leery of going ahead and completing the meta data for all the files and then coming back to track 2, just in case something weird happens.
I just scrolled down to Track 15 in Metadata. By the time I reached number 15, three tracks had ‘dropped out’. I haven’t gone on to number 70, but at this rate, if I had I am sure there would be many more missing. As I said in the last post, all these tracks are numbered sequentially, starting at 201 and ending at 270.
