Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture explains the challenges of analyzing speech signals due to the varying spectral content of vowels and consonants. It demonstrates how spectrograms help in splitting speech into segments for analysis, showcasing the differences in wideband and narrowband spectrograms. The lecture also delves into the concept of short time Fourier transform, illustrating how it leads to a uniform tiling of the time-frequency plane and the implications of the uncertainty principle in time-frequency analysis.