Top 8 Human Speech Emotion detection or recognition using deep learning algorithm

Top 8 Human Speech emotion detection is identified in this project. It useful to understand the human emotions when they deliver through speech. Emotion of 8 categories namely Calm, Happy, Sad, Angry, Fearful, Disgust, Surprised and Neutral. These emotions are considered with female and male audio “.wav” wave files are considered as dataset. The speech emotion detection dataset represents as follows

File name example 03-01-01-01-01-01-02

Audio-only (03)
Speech (01)
Neutral (01)
Normal intensity (01)
Statement “dogs” (02)
1st Repetition (01)
12th Actor (02)
Female, as the actor ID number is even

The dataset is trained with Convolutional Neural Networks (CNN) algorithm. More specifically, CNN1 is used for training and validation. The implementation includes live detection of emotion from audio file, some audio we tested from Youtube. There are 8 emotion is classified in this project. Speech emotion detection is more useful to identify the person’s emotion without seeing face. The source code is neatly written in Python for speech emotion recognition from audio files. The following image represents the result of emotion on real time detection from audio input.

You can Check the demo video below with complete implementation

Human emotions are of two types, speech and face emotion. Here we studied Speech emotion. Human face emotions are also discussed in our previous post, Check the link

4 thoughts on “Speech Emotion detection using Deep learning

Leave a Reply

Your email address will not be published. Required fields are marked *