Emotion Identification is a machine learning project that aims to classify emotional states based on audio recordings. This project processes and analyses the data using a variety of Python libraries, including librosa, numpy, pandas, seaborn, matplotlob, and sklearn. The CREMA-D dataset is a collection of audio recordings of actors expressing various emotions. The project makes use of librosa to extract audio properties like pitch and energy, numpy and pandas to analyse and manipulate the data, seaborn and matplotlib to visualise the data, and sklearn to create machine learning models. This project's ultimate objective is to correctly determine a speaker's emotional state from their audio recording.
The CREMA-D dataset consists of audio and video recordings of 7,442 different people displaying various emotions. It is a useful tool for study on the identification and recognition of emotions since each clip is tagged with labels indicating the main emotion being exhibited. The dataset is well-balanced, with representation from a wide variety of ages, genders, and races. It is excellent for developing and testing machine learning models for problems involving emotion recognition.
To use this model, you will need to have the following dependencies installed:
- numpy
- pandas
- matplotlib
- seaborn
- sklearn
- librosa
 |
 |
 |
 |
|---|---|---|---|
| Nayana Krishna | Skandhan R Nair | Kristiena Benny | Joel Georgie Jacob |