L541: Introductory Phonetics (Spring 2004)
[General Information | Weekly Class Schedule | Lab Information | Links]

WaveSurfer Guide for L541

Part 6: Spectrogram

[Index | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 |10]

Preparation: Use the sound file hw06_boston.wav for your practice. Open the file with the spectrogram cofiguration.

In order to get the desired time/frequency resolution of the spectrogram, you can change several parameters in the Properties of Spectrogram (Right-click on the spectrogram and select properties).

Parameters of Spectrogram Properties

FFT window length: should be a power of 2 (256, 512, 1024, etc) for fastest computation of the spectrogram. Use a large number when sampling rate is high. When you click Sound tab in the Properties window, you can find out the sampling rate of the sound file.

Analysis window type: Use Hamming window.

Analysis bandwidth: There are two kinds of spectrogram analysis: narrowband spectrogram and wideband spectrogram. In general, wideband spectrograms are used for phonetic studies. A bandwidth used for wideband spectrograms is 300-500 Hz. A bandwidth for narrowband spectrograms is 30-45 Hz.

Pre-emphasis factor: Use default value. Avoid 1.

Cut spectrogram at: Maximum frequency to display. Use default value or 8000. If this value is higher than the Nyquist frequency of the sound (half of its sampling frequency), some values in the spectrogram will be zero, and the higher frequencies will be drawn in white.

Brightness and Contrast: Click the Controls button. The following window shows up. Move the cursor to adjust the color resolution.

Grid frequency spacing: Use default (= 0 Hz). If you want to add grid on the spectrogram, change this value.

Grid time spacing: Use default ( = 0 s).

Grid color: Use default ( = red) or any color.

Spectrogram color: Use grey

Prepared by Kyoko Nagao
Last updated on Feb 28, 2004