Topic: Timestrech

Viewing 3 posts - 1 through 3 (of 3 total)
  • May 7, 2010 at 03:52 #2222
    Lion
    Participant

    Any way we’ll be getting this?

    I *was* able to do some timestreching in a song I know I want to do for my album. In shortcircuit lowering the pitch (which, of course, slows it down), and then using zPitch to raise the pitch back to the original (3 semitones down SC, 3 semitones up on zPitch). TBH, I didn’t even realize “Hey, I’m timestreching!” ’til after I did it. lol.

    But when it comes to the stretching I (and I know any other ppl that used other progtams) want is to have timestreching with the audio on the timeline. I know theres the beatslicing>change to desired tempo>and merge, butttttt this is a bit cumbersome.

    I dont know much about Fourier transforms and all that, but isn’t time streching and pitch shifting basically using the same methods?

    May 7, 2010 at 08:25 #18021
    kyran
    Participant

    @Lion wrote:

    I don’t know much about Fourier transforms and all that, but isn’t time stretching and pitch shifting basically using the same methods?

    Time stretching is actually done using some sort of granular processing: you slice the audio very very very thin and then move them away from or closer to each other. (this is a very simple visualization of the process)
    It’s notoriously difficult to get it to sound good though.
    All time stretching options you find in other programs are basically the same algorithm licensed from zplane.

    btw: you’ll not often find real time audio things working directly with frequency data. The most efficient way to perform a Fourier transform is the FFT algorithm, which processes data in blocks. You can’t just use the FFT on one sample. Processing data is blocks means lots of latency. It’s the main reason almost all convolutional reverbs add latency.

    May 7, 2010 at 12:28 #18024
    Zynewave
    Keymaster

    @Lion wrote:

    Any way we’ll be getting this?

    Yes. I’m still gathering strength for starting work on it :wink:. I don’t have an estimate yet, as I’d like to have all minor stuff I have on my todo list dealt with before I start on this.

    I dont know much about Fourier transforms and all that, but isn’t time streching and pitch shifting basically using the same methods?

    I made zPitch actually to get the technology that I need when eventually implementing time-stretching. The pitch shifting is achieved by sample rate converting the sound and then time-stretching it so that the speed is the same when played back at the original sample rate.

Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.
© 2021 Zynewave