I downloaded and listened to the audiobook This is Our Music: Free Jazz, the Sixties, and American Culture. The author is Iain Anderson, and the language and structure seem like those of a slightly rewritten doctoral dissertation. It’s pretty interesting as a source for the politics — particularly racial politics — of jazz in the late 50s and early 60s, and it held my interest for the 5 hours I needed to listen to it at double speed. But what really fascinated me was the reader’s voice. The reader is listed as Paul Steven Forrest, but I can hardly believe that this is a human voice. (Indeed, this is the only book that this name has been assigned to as reader.) The sentence intonations are much too regular, and seem to ignore any cues related to the meanings of words. Some reasonably common English words — at least, common enough in academic jargon — such as “diaspora” are systematically mispronounced, but without any hesitation such as you might expect from a human reader stumbling over an unfamiliar word. Similarly, non-English words were completely botched, but without apparent self-consciousness.
On the other hand, if Paul Steven Forrest is in truth the pseudonym for a computer-generated voice, it’s remarkably good, at least to someone who has not been following progress in speech generation over the past decade. It took me an hour of listening before it struck me that something was off about the voice, and while it started to bug me, it never became unbearable.