Skip to Content

Department of Linguistics

SPEECH ACOUSTICS

Frequency Transformations of Speech

Robert Mannell

Frequency domain psychoacoustics is complex and this complexity is due in part to two parallel methods of processing frequency. These two methods of frequency processing are known as the "resonance" or "place" mechanism and the "frequency" mechanism.

The place mechanism is based on the resonance characteristics of the basilar membrane in the inner ear. Each point along the length of the basilar membrane moves most strongly in response to a specific frequency, its characteristic frequency (CF). The inner hair cell (sensory cell) at each position along the basilar membrane shares the CF of that part of the basilar membrane. Each hair cell responds best to its CF and responds less strongly to a range of frequencies on either side of the CF. As we move from the basal end (near the oval window) to the apical end of the basilar membrane, CFs change progressively from high frequencies to low frequencies. The frequency mechanism refers to the rate of firing of individual auditory nerve fibres (which are stimulated to fire by the inner hair cells). Nerve fibre firing rate directly encodes the lower frequency components of a sound. The place mechanism is responsible for the frequency selectivity (resolution) of the ear whilst the frequency mechanism is responsible for the perception of pitch (accurate up to about 1000 Hz, and less accurate up to 3000 - 5000 Hz). For more information on these mechanisms see the web page on "Theories of Hearing".

1. Pitch Perception

Click here to view this topic.

2. Frequency Discrimination

Click here to view this topic.

3. Frequency Selectivity

Click here to view this topic.