Unknown component type error in emotion recognition

[nancysemwal]

Following this tutorial for version 2.1.0 for performing emotion recognition, I copied the folder models into my opensmile top directory from here (the link doesn't work anymore by the way). Everything in the modles folder is exactly like the tutorial says.

However, I get the following error message on using the emobase_live4_batch_single.conf configuration file:

SMILExtract -C config/emobase_live4_batch_single.conf -I input.wav > result.txt

(ERROR) [1] in cComponentManager : cannot add component (instname='valence' type='cLibsvmLiveSink'): unknown component type!!

I understand what the error implies but I am unable to figure out what can be done to eliminate it. Also there is an interesting part. I had downloaded the models folder from the link around 1-2 months back, but didn't run the emotion recognition example at that time. When I went to the link again today, the folders are no more available and I get a 404 error.

I get the same error for a different component while performing Voice Activity Detection (VAD) on using vad_segmenter.conf config file:

(ERROR) [1] in cComponentManager : cannot add component (instname='lstmVAD' type='cRnnProcessor'): unknown component type!!

Does anybody knows whats happening? Is this because the tutorial was for version 2.1.0 and hence the models folder would be for that version, while I'm using version 2.3.0?

[harshulgupta2]

Hi, Even I was trying to run Open Smile and I was unable to get the model files. Can I get some help in this regard.

Thanks