Explore the intricacies of Mel-Frequency Cepstral Coefficients (MFCCs) in this comprehensive 58-minute video tutorial. Delve into the concept of Cepstrum, its intuition, and the process of extracting MFCCs, which are widely used in speech and music processing. Learn about the historical context of Cepstrum, visualization techniques, and the role of MFCCs in understanding the vocal tract and speech generation. Discover how to compute and visualize MFCCs, understand their applications and limitations, and gain insights into speech formalization and component separation. Access accompanying slides for enhanced learning and join a community of AI enthusiasts for further discussions and networking opportunities.
Overview
Syllabus
Intro
Join the community!
An historical note on Cepstrum
Computing the cepstrum
Visualising the cepstrum
The vocal tract
Speech generation
Understanding the cepstrum
Formalising speech
The goal: Separating components
Computing Mel-Frequency Cepstral Coefficients
Why Discrete Cosine Transform?
How many coefficients?
Visualising MFCCS
MFCCs disadvantages
MFCCs applications
What's up next?
Taught by
Valerio Velardo - The Sound of AI
