#fft #nlp


A tool to process speech audio in Rust

7 releases

Uses old Rust 2015

0.3.0 Dec 31, 2017
0.2.4 May 29, 2017
0.2.3 Sep 24, 2016
0.2.1 Aug 4, 2016
0.1.0 Feb 2, 2016

#506 in Audio

32 downloads per month

MIT license

1.5K SLoC

Build Status


A tool to let you hack away at voice audio recordings in Rust.

How do I do it?


What's included

  • Filter preemphasis, normalization, RMS calculation
  • Autocorrleation calculation
  • Laguerre root finding and polynomial division
  • Mel-Frequency Cepstral Coefficient (MFCC) calculation
  • Linear Predictive Coding (LPC) coefficient calculation
  • Formant path finder (McCandless algorithm, from Praat)
  • Pitch finding (Boersma autocorrelation method, from Praat)

Why is it broken?

Open an issue! There are many incomplete aspects to this library, as it is highly specialized to my personal projects. Once others begin using it, I expect that it will become more generalized.


~59K SLoC