Hello all,
I am interested to learn about vocal emotion recognition softwares (for animals). I’m aware of LIVEQUEST for chickens, and SOUNDWELL for pigs. As far as I know, the softwares I mentioned are still being built in laboratory, and then they are not available to
use. So, I came across with the speech emotion recognition softwares like this on YT: https://youtu.be/NQ2-kVNIaF0. However, it is to voice analysis only.
My interest is to find a software that be able to build recognizers from 11 frequency parameters approximately. The reason for this is because animal vocalizations are usually studied with many parameters to tell about emotions.
For example:
- duration,
- duration from start to maximum amplitude,
- interval (distance to previous call),
- peak frequency,
- the minimum and maximum frequency and resulting bandwidth,
- three quartiles that describe the distribution of energy over the frequency range (Q25 refers to 25%; Q50, to 50%;
and Q75, to 75% percent of the energy in the frequency range),
- the number of peaks (above −20 dB, hysteresis: 10 dB),
- the frequency of the first two peaks (F1 and F2),
- the entropy and the harmonic-to-noise ratio (HNR)
I’d appreciate your recommendations on softwares and laboratories developing vocal emotion recognition softwares.
Kind regards,
Aline
Sent from my iPad
|