Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Real-Time Contrast Enhancement to Improve Speech Recognition

Figure 1

Schematic of the contrast enhancement algorithm.

The incoming signal, x(t), is first decomposed into 110.25-Hz wide frequency channels using polyphase decomposition (see text). Intermediate processing incorporates a winner-take-all (WTA) strategy in which channel gain is weighted in a way that simulates a lateral inhibitory network and sharpens spectral contrast. The output signal, y(t), is synthesized by an inverse of the process used to analyze the signal.

Figure 1