In fairness the two terms are used rather interchangeably, and often confusingly often by people who by rights should know better.
For example, Win frotz tts was a preconfigured version of win frotz which had sapi output enabled automatically, it was also the one Howard sherman packaged with several of his stand alone interactive fiction titles supposedly aimed at blind people.
However, regular winfrotz had a "speech" function which pretty much did the same thing, aside from the fact that any and all screen readers could read the screen of winfrotz with a virtual curser, and some like NVDA could directly speak output anyway.
So essentially someone constructed a supposedly "accessible" tts version of a program that was accessible enough in it's default state anyway, and even had an in built speech function, yet it was assumed that blind people would need the tts.
On the other hand, when Jim Kitchen bought out a new version of mac 1 which ran directly through his main winkit program rather than being stand alone, it was quite logical for him to describe it as "mac 1 tts", given that original mac 1 had used pre-recorded speech, whilst all the games which ran with Winkit by default used microsoft Sapi for output.
Just to be even confusinger, I've recently also seen developers using the term "voice over" for programs on windows, when they actually mean either tts or self voicing.
Btw, hmmm, wonder if we should construct an article in the articles room which defines these things?
With our dreaming and singing, Ceaseless and sorrowless we! The glory about us clinging Of the glorious futures we see,
Our souls with high music ringing; O men! It must ever be
That we dwell in our dreaming and singing, A little apart from ye. (Arthur O'Shaughnessy 1873.)