A colleague and I are teaching a Voice Disorders class, and want to give the students experience with acoustic measurement of dysphonia, such as CPP and HNR. What's the best way to do that? Can it be done with Praat?
These two acoustic measurements can be calculated with the praat. The CPPs are included in the praat in the latest versions. This cepstral measure can also be calculated with the speech tool program. Nearly perfect correlation has been found in the CPPs calculated between the two programs, although they can not be compared numerically due to differences in signal processing [Maryn, Y., & Weenink, D. (2015). Objective dysphonia measures in the program Praat: smoothed cepstral peak prominence and acoustic voice quality index. Journal of Voice, 29(1), 35-43]
A sustained vowel can be used to calculate the HNR.
Both sustained vowels and connected speech can be used to calculate CPPs.
The AVQI (Acoustic Voice Quality Index) uses the combination of a sustained vowel with the connected speech. With this index is calculated, among others the CPPs and the HNR. [Barsties, B., & Maryn, Y. (2016). External validation of the Acoustic Voice Quality Index version 03.01 with extended representativity. Annals of Otology, Rhinology & Laryngology, 125(7), 571-583]
Regards,
Jonathan
This can also be done with VoiceSauce: http://www.seas.ucla.edu/spapl/voicesauce/. If you are aware of Praat vs VoiceSauce differences in the calculations of CPP and other voice source measures, I'd be very interested to learn about them.
VoiceSauce is certainly very user friendly.
Thank you, Misa. I do not know voice sauce, but will have some of my students try it.
Goodmorning, do you know how to derive the cpp value from the praat?
Would this help at least a little bit? http://www.fon.hum.uva.nl/praat/manual/PowerCepstrum__Get_peak_prominence___.html
- Badges
- Science topic
- Similar topics
- Correction