So the manual states:
The Cakewalk TTS-1 soft synth supports 44.1 kHz, 48 kHz, and 96 kHz sample rates,
but not 88.2 kHz. The audio engine will not engage if SONAR is configured to use a sample rate of 88.2 kHz
while TTS-1 is loaded and connected.

Well as we all know by now contrary to the manual, the TTS-1 in fact is distorted at 96 kHz. However there is a
workaround by opening the TTS -1 System Settings/Options/Options/Check the box "Light Load Mode" and voilà.
The TTS-1 then sounds perfect. The issue now requires applying the workaround every time a project is opened.
The workaround cannot be saved. A project containing several instances of TTS -1 requires applying the workaround
to each TTS -1. The developers could add a save tab. The TTS -1 is quite useful and worth updating. Any ideas?

Good to know.
 
With one exception, I've never been able to get anything from TTS-1 except completely garbled noise, so I've disabled it entirely from my plug-in searches and options. The one exception to this garbled noise rule is when I switched the audio engine to run in 32-bit mode instead of 64.
Well, what good is having a 64-bit PC with a 64-bit DAW if you have to run in 32-bit for one plug-in?

I recommend enabling TTS-1 and using the "Light Load Mode". It does work quite well at 96 kHz.
I have requested that Cakewalk developers provide a reasonable solution to keep the feature
"Light Load Mode" saved when enabled. We'll see how that goes. Best wishes.

David R.
So the manual states:
The Cakewalk TTS-1 soft synth supports 44.1 kHz, 48 kHz, and 96 kHz sample rates,
but not 88.2 kHz. The audio engine will not engage if SONAR is configured to use a sample rate of 88.2 kHz
while TTS-1 is loaded and connected.

Well as we all know by now contrary to the manual, the TTS-1 in fact is distorted at 96 kHz. However there is a
workaround by opening the TTS -1 System Settings/Options/Options/Check the box "Light Load Mode" and voilà.
The TTS-1 then sounds perfect. The issue now requires applying the workaround every time a project is opened.
The workaround cannot be saved. A project containing several instances of TTS -1 requires applying the workaround
to each TTS -1. The developers could add a save tab. The TTS -1 is quite useful and worth updating. Any ideas?

Thanks for this tip. I could never use it because I use 96k. I just tested it and it works. Thanks again.