Syllabic compression: Effective compression ratios for signals modulated at different rates

Compression circuits are being used increasingly in hearing aids to reduce the dynamic range of signals. Their performance is usually characterized by: (1) the threshold sound level above which the compression starts to operate; (2) the compression ratio, which is the change in input level (in dB) required to achieve a 1 dB change in output level; and (3) the attack and release times over which the signal is integrated to determine the necessary gain change. In many practical situations, the effective compression ratio obtained with dynamically varying signals such as speech is less than the compression ratio obtained using standard test signals (slow square-wave modulation with large modulation depth). This article describes the effective compression ratios achieved with sinusoidal modulation, as a function of modulation rate, level relative to the compression threshold, compression ratio and time constants. The effects of compression on a typical speech signal are also discussed.