Take the 2-minute tour ×
Sound Design Stack Exchange is a question and answer site for sound engineers, producers, editors, and enthusiasts. It's 100% free, no registration required.

https://soundcloud.com/drew-blaisdell/low-sample-test

I recorded a note on my guitar into Logic Pro. This is the first note you hear played in the clip above. It sounds pretty much like a note on my guitar.

The second note is the same sample put through Logic's Time and Pitch Machine to make it an octave lower (-1200 cents) while keeping the length of the sample the same. It is a bit wobbly/choppy.

The third note is the second sample lowered an additional six tones (-600 cents). It is also distorted/wobbly/choppy.

Why does this happen? I (wrongly) assumed that simply lowering the frequency of a guitar sample while keeping the period the same would make it sound like a bass guitar. Are there better ways of pitch-shifting a sample?

share|improve this question

migrated from avp.stackexchange.com Jan 24 at 12:01

This question came from our site for engineers, producers, editors, and enthusiasts spanning the fields of video, and media creation.

1 Answer 1

up vote 6 down vote accepted

The only "natural" (i.e. "using only the data in the sample") way to change the pitch of a recorded sample is to change the speed of its playback. So a sample that sounds an octave lower should take twice as long to play back.

But you have a tool that somehow manages to play that sample back octave lower, while somehow keeping the playback time the same. How can this be? It is achieved through clever algorithmic modification of the sample's playback. However, this modification is a distortion of what it would "normally" sound like, and you get these undesirable extra sounds as a consequence. In general, these extras are called "artifacts."

The good news is that while you can't avoid artifacts entirely, you can optimize for different scenarios. There are many techniques for pitch and time modification, based on concepts like chopping the sample up temporablly into small loopable slices, or deconstructing it harmonically, and reassembling the transformed parts into a new recording. Many tools let you choose from several algorithms, with different tradeoffs. For example, chopping up over time preserves transients, which is very nice for percussion and drums where the transient "hits" are more prominent than the decays, but is problematic for tonal sounds where smoothness is more important (since you can hear the slices).

You may want to look into your Time and Pitch Machine and see if you can choose a different algorithm that might be better suited to your sample.

share|improve this answer
    
+1 And there's one thing to add - pitch is only one of the numerous sound qualities. Just algorithmically modifying a sample's pitch will not turn a violin into a cello. Even when dealing with single instruments, the sound generated by most of them is so complex that any stretching greater than couple semitones (at most) will sound fake. –  Karol Piczak Jun 25 '13 at 21:27

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.