I'm looking for a way to extend certain parts of an exiting audio file to make these parts sound like the where played longer. And I dont want to change play speed or stretch the file. Lets say u have a digital sound file of someone playing a trumpet. And he's playing diffrent tones. Now I want to jump to a certain point in this file and hold this very tone as long as I want. To make it sound like it was played very very long. Without anyone noticing that it was artificialy extended. And I want to do this with not just recordings of instruments.

What technics do you suggest?
Maybe something like granular synthesis or any other synthesis technic?

It would be so awesome if someone could help me with this.
Thanks in advance


1 Answer 1


Sounds for me, like non-linear time stretching can accomplish what you want. In your example, you would apply a stretching function with a slope of one most of the time and for the single note (you want to be played very long) the slope then needs to be strongly different from one.

Overview paper: Jonathan Driedger and Meinard Müller. A Review on Time-Scale Modification of Music Signals, Applied Sciences, 6(2): 57–82, 2016. [DOI]

Code: https://www.audiolabs-erlangen.de/resources/MIR/TSMtoolbox

  • $\begingroup$ That sounds promising. Thank you very much $\endgroup$
    – cube
    Commented Mar 11, 2019 at 16:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.