Skip to content

Latest commit

 

History

History

README.md

LAB02: SPECTROGRAMS

Tutorials

  1. Spectrogram ( HTML )

shows how a spectrogram is the result of the sliding window appoach and the computation of short-time Fourier Spectra.

  1. MelScale ( HTML )

discusses the design of the mel scale in a number of different variants. It also shows how to design a mel scaled filterbank that is used to convert a Fourier Spectrogram into a mel spectrogram

  1. MelSpectrogram ( HTML )

shows how we transform a traditional Fourier spectrogram into a mel spectrogram using mel scaled filterbanks.

Exercises

FilteredSignals.ipynb :

  • Filt-Ex.1: Speech with Telephone Bandwidth
  • Filt-Ex.2: Redundancy in bandpass-filtered Speech
  • Filt-Ex.3: Filtering of Harmonic Signals

Filtering an audio signal, i.e. removing part of the frequency spectrum, has an immediate impact on audio quality. At the same time its impact on speech understanding can be small thanks to redundancy in the speech signal. A classical example is speech passed over a telephone line. These exercises go into more detail and elaborate on the links between spectrum and perception.

SpectrogramGUI.ipynb :

  • Spec-Ex.1: Phonetic Segmentations
  • Spec-Ex.2: Fourier Spectrogram: parameters

(Ex.1) Overlaying phonetic segmentations on a speech spectrogram illustrates the discrepancy that exists between the continuity of the speech signal and the discreteness of any symbolic (graphemic or phonetic) representation. Phonetic boundaries tend to be fluid rather than abrupt.

(Ex.2) A spectrogram is a time-frequency representation. However, different parameter settings may lead to different interpretations of what we see in the spectrogram. As a default we use established default parameters that are in accordance with the human auditory system and that work well in speech processing.
However, different parameter settings give different tradeoffs and may give a very different 'view'. Discover for yourself the Heisenberg uncertainty principle of spectral analysis and find those parameters that best aligns the spectrographic view with perception in general.

MelSpectrogram.ipynb :

  • MelSpec-Ex.1: Mel Spectrogram - Answer the questions at the end of the Tutorial notebook