After decades mixing live sound, I've plunged into audio books, and besides mouth clicks, balancing voice spectrally has emerged.
The spectral view of the current voice recording looks normal up to 3k, then plunges about 9db to around 4k, then rises about 6db in a smooth hump to around 6-7k and back down to normal around 10k. I recorded myself and another, both of which were smoothly distributed, verifying the "abnormal" spectral distribution was just my narrator's unique voice pattern, not the mic/recording system.
Question - should I leave it as is, respecting this unique voice pattern, or e.q. to "normalize" to a smooth spectral upper decline, taking advantage of the increased intelligibility around 4k and smoothed out sibilance and mouth clicks around 6-7k?