Skip to content

CNN for language-independent emotion prediction on a dataset of English and French speech samples.

Notifications You must be signed in to change notification settings

kaplanan/deep-emotion-sense

Repository files navigation

deep-emotion-sense

Convolutional neural network to predict language-independent emotion on a dataset of English and French speech.

This model is trained on a merged dataset of natural speech with data provided by IEMOCAP for english speech samples and Recola for french speech samples.

The implementation is a basis for a research paper on comparison on different optimization algorithms and activation functions.

Getting Started

You can view the notebook here.

Run the notebook

Prerequisites

  • Python 3
  • Tensorflow

Starting the notebook

Simply open a new terminal in the directory and type:

> jupyter notebook

Built With

Contributors

About

CNN for language-independent emotion prediction on a dataset of English and French speech samples.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages