Recognize Speech Emotions With Librosa & JupyterLab - Workshop
Recognize Speech Emotions With Librosa & JupyterLab by The Assembly
The human voice conveys underlying emotion through tone and pitch while producing speech. While deciphering this can be extremely subjective, the field of Speech Emotion Recognition (SER) aims to do this autonomously using advanced algorithms and machine learning.
In this workshop, we will use Librosa , a Python library for audio analysis, with a model trained on the publicly available RAVDESS (Ryerson Audio-Visual Database of Emotional Speech and Song) dataset, to achieve SER and identify emotion features from speech. We’ll run our code using JupyterLab - an open source, web-based UI for Project Jupyter - which allows for step by step execution of Python command batches with interim result display (similar to Jupyter Notebooks), so that we can examine the process in greater detail.
- Bring your laptop
- Some programming and data science knowledge
This workshop is free admission.