Speech Analysis and Voice Identification at the Forensic Media Lab™ includes the generation of voiceprints to examine human speech. Voiceprints show the frequency components, or pitch levels, of speech sounds and the corresponding amplitudes, or volume levels, of the frequency components as they occur over time. A voiceprint thus makes it possible to visually determine the speech characteristics of a particular person. Here are examples of voiceprints depicted through a Spectrogram, also sometimes called a Sonogram:

 

 

 

 

In the Spectrogram, the horizontal axis represents time while the vertical axis represents frequency. The amplitude of each frequency component is represented by the color intensity of each point in the voiceprint. A Spectrogram is very useful for identifying distinctive spectral patterns created by human speech.