I’m recording a cassette tape from the year 1979 with the words “Mono” on it. Audacity automatically creates a stereo track when recording starts. I guess this is because the tape is played back in a stereo cassette player and my PowerBook receives a stereo signal through its line-in port. I have verified that the waveforms in the left and right channels of the stereo track are identical. Now I’m wondering If I should keep two channels or not with regard to (1) file size and (2) what media players prefer.
An uncompressed, lossless file made from a single mono track does no have any apparent difference to a file mixed down from both mono tracks (made by splitting the stereo track). But the stereo track has about twice the file size.
What about media players? Is there an audible difference between a mono track sent to both left and right speakers compared to sending seperate channels (with identical audio)? I can discern a clear difference - the stereo version is loader - but why is that?
Theory has it that a Mono music file properly tagged should play on most if not all music players and present audio from both speakers. There should be no difference between that and playing the same stereo file with the exact same work on both tracks.
However, if you’re worried about this, capture and archive/store in mono and copy/link that one track to two-track stereo before you use it for any production.
I can discern a clear difference - the stereo version is loader - but why is that?
This is where you get to roll up your sleeves and tell us exactly the conditions where you found that effect. You posted in a portion of the forum that tells us nothing about you. Computer? OS? Audacity Version? Players? Model numbers?
As I said, the stereo.aiff sounds loader, i.e. more accentuated all over.
I’m recording to a PowerBook running Audacity is 1.3.14 (Beta) on OS X 10.4.11. I used a pair of well-shaped 30-something ears to compare the audio in QuickTime Player Pro 7.6.4 (518.35).
Koz, I’m not sure if you rather mean channel when you write track. Why should I capture onto a mono track to then duplicate that. Does an export created from two identical tracks differ from one stereo track with identical audio on either channel?
The tape is monophonic recording. Played on a stereo player theoretically should give a dual-mono result: two perfectly identical tracks, but in reality the tracks are not truly identical* : the frequency content (equalization) is different for each track, creating a pseudo-stereo effect. If you make a truly mono version you can hear the difference in the stereo image …
[*Try splitting the stereo to mono and inverting one track, if they were truly identical you should then hear nothing : total destructive interference ]
If the 2 channel track was mono then they would (or at least “should”) sound identical, however your “stereo.aiff” file is not mono.
Here I’ve zoomed in very close on the track and you can see there is a phase shift between left and right channels (the right channel is delayed by 10 samples).
Wow, interesting! I didn’t notice the subtle differences in the waveforms before I opened this thread.
So, the stereo version does have more sound information. But it actually sounds a little too strong to my ears whereas the mono version sounds more natural. I guess this indicates that I should adjut input volume in Audacity or play trim on the cassette deck?!
Thinking out load, how does a mixed down mono version created from mono tracks (former left and right channels of the captured recording) compare to the unaltered stereo version? Shouldn’t they contain the same sound information… so its only the separation (of the same audio) into channels that creates richer sound to the ears…hm, I have some learning to do…
And by the way, does Audacity have a build-in tool (algorithm) specifically for creating a pseudo-stereo effect from a true mono source, as Trebor wrote about?
Because of the phase differences between the channels, that will probably be a better way to convert to mono than mixing the channels to mono (Tracks > Stereo Track to Mono). Mixing the channels to mono will cause “comb filtering” and will cause frequencies at odd multiples of 2205 Hz to be attenuated (filtered out).
Sample rate 44100 Hz
10 sample shift = 10/44100 = 1/4410 second (delay length)
Notches occur at 1/2K, 3/2K, 5/2K … where “K” is the delay
which gives us notches at:
2205 Hz, 6615 Hz, 11025 Hz, …