Nueral Network for Emotion Classification

I'm seeking a freelancer experienced in building machine learning models. Data is parsed and stored in arrays format as follow:

1) Audio features using librosa
2) Spectrogram images

The task entitles creating 8 models as follow:
1) two different audio classification models
2) two different image classification models
3) 1 ensmeble model from the two audio models
4) 1 ensmeble model from the image models
5) 2 multimodals (early fusuon and late fusion) from audio and image modities

Data is imbalanced, 6 emotions

Reach out to me ONLY if you have strong experience in building audio and image classification models.

You will build/fine tune the models. Data processing is done