I'm trying find authoritative references for
 Which analyzing techniques (time/frequency domain) i should use for determining the quality of a speech signal.
 Which algorithms i should run for improving speech signals( As "cheaply" as possible)
There are no constraints for the first requirement as it will be done "offline". But I'm inteding to do the second "on the fly", So i need algorithms which will do its job as economically as possible. In the second case, dedicated DSP hardware may or may not be available. I've to explore both options.