Hello everyone I'm working in audio analysis for emotion classification. I'm using parselmouth (a PRAAT integration in python) to get feature. I'm not well versed in audio analysis, starter. After reading many papers and forums. I see mfcc are used for this, I've also discovered some features they're (jitter, shimmer, hnr, f0, zero_crossing) are they used for this work?
What I've to do with audio files before extracting mfcc and these features?
After getting these features I've to predict emotion using machine learning.
It'll involve:
- The algorithm must be able to make predictions in real time or near
- Taking into account the sex and the neutral voice of each person (for example, by reducing and centering the variables of the model to consider only their variations with respect to the mean - average which will thus change value as and when the sequential analysis since it will be first calculated between 0 and 1 second, then 0 and 2 seconds, etc.)
Any help and suggestion for best practice are welcome.
Thanks
Here are some papers that might be helpful to your work:
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0062397
http://www.homepages.ucl.ac.uk/~uclyyix/yispapers/Xu_Kelly_Smillie2013_author_version.pdf
For a tool that can extract emotion-relevant measurements, please see ProsodyPro: http://www.homepages.ucl.ac.uk/~uclyyix/ProsodyPro/
Here are some papers that might be helpful to your work:
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0062397
http://www.homepages.ucl.ac.uk/~uclyyix/yispapers/Xu_Kelly_Smillie2013_author_version.pdf
For a tool that can extract emotion-relevant measurements, please see ProsodyPro: http://www.homepages.ucl.ac.uk/~uclyyix/ProsodyPro/
Hi Abdoulaye Diallo,
You should make sure you read several other papers in the field to see what others are doing. The techniques and approaches will vary slightly, depending upon the type of audio you are working with: speech, music, singing, effects, etc.
The features you mention are quite commonly used. You should also look at others like energy and spectral centroid. They are usually helpful too. Check out the MIR Toolbox (https://www.jyu.fi/hytk/fi/laitokset/mutku/en/research/materials/mirtoolbox) and Matlab Audio Analysis (https://www.mathworks.com/matlabcentral/fileexchange/45831-matlab-audio-analysis-library) libraries too for an idea of other features.
Before extracting features, you may wish to consider adjusting the sampling rate for your filed (for consistency) and peak-normalising the files - although you should be comfortable that you understand the reasons for doing this.
On the recognition of emotions from audio recordings.
It is necessary to identify the measurable cognitive precursors that can induce emotion, they are essential since the innocence in front of the stimulus that faces the subject carries the emotional potentiality.
To be or not to be, that's the question: Sakespere said in the famous monologue phrase of the main character Hamlet.
As for the emotions, it could be added:
Knowing or not knowing, that's the emotion.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python