I'm working on rolling noise emissions from cars. I have a model for rolling noise emissions which gives me sound pressure levels in third octave bands (29 in total, between 20Hz and 8kHz), depending on vehicle speed, and road/tyre combinaison.

I'd like to fit a real recording of tire/road noise to the model, while keeping the spectral properties of the recording. The signal of rolling noise is mainly stochasticenter image description here

In blue, the spectrum of the recording using Welch's method, and in red, the levels i'd like to reach.

What kind of methods can i use in this particular case?

Thank you for your help!

  • 1
    $\begingroup$ I don't really understand – usually, you fit the model to the data, not the data to the model. What is it that you want to achieve? $\endgroup$ – Marcus Müller Mar 10 '17 at 15:45
  • $\begingroup$ So you want your recording to sound like your model? Why don't you apply a multiband equalizer? You already know the desired gain in dB for each band. $\endgroup$ – Juancho Mar 10 '17 at 19:07
  • $\begingroup$ I want a realistic sound of rolling tire noise, that's why i use a recording, and i want to fit it to the true levels given by the spectrum in red. $\endgroup$ – Edward Mar 11 '17 at 22:24

protected by jojek Mar 13 '17 at 8:38

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Browse other questions tagged or ask your own question.