A computer-implemented method performed by an electronic data processing apparatus to implement a suggestion engine, comprising: configuring a variable, which is independent from an attribute associated with content of a digital audio recording featuring one or more human voices, by: defining a quality score impact function characterizing a plurality of weight values having a maximum in a central band and tapering away in one or two respective bands adjacent to the central band maximum to produce a range of weight values, and defining a scale setting a minimum and a maximum range of values for the variable corresponding to the range of weight values.