The ideal of polite conversion is that people who disagree can come to a conclusion, a "truth," that was previously inaccessible to either of them. Asserting "the truth," is a dominator paradigm privilege.
And yet, in your opening paragraph, you described "the truth". "The truth" would be that a percentage of population complains about a certain sound while the remainder of the population doesn't. You might even be able to describe physical properties of that sound, such as the range of frequencies and the amplitude that produces discomfort in a given proportion of population. You might even be able to describe the characteristics of populations that respond to the sound differently (such as, I don't know, socioeconomic status for whatever reason). That would probably bring you as close to "the truth" as possible. Whether the sound in question is called "high-pitched" or not is irrelevant.
It irks me that subjective claims about the physical world are regarded as truth statements about the physical world, and thus my report that I hear a displeasing sound and your report that you don't hear that sound or don't find it displeasing somehow produces two "truths".
And if we're talking about something that isn't physical, I suggest that it's still the same. If there is no objective correlate, then... let people have their opinions, but don't label either side "truth".
How is the power of the speaker relevant to the truth of what they are saying?
Would the presence of a high pitched sound change if the people in power say there /is/ a high-pitched sound, and the people out of power say it doesn't exist?
The way we find truth is to measure the sound, rather than to pay undue attention to anecdotal evidence to one party based on identity.
I found this such a bizarre analogy (responding to pg's point, not directly to you). There can be meaningful difference of opinion and perception on whether or not a sound exists; people literally might not have the physiological equipment to hear it. However, the assertion that "there are things I can't say" (meaning, presumably, "there are things I will experience significant hardship for saying" rather than literally "… things I am physically incapable of saying") is surely proveable by mentioning such a thing. He dismisses this idea, but seems never to explain why he's dismissing it, other than by this analogy, which to me seems false.
If the answer to why you can't say what things can't be said is that merely specifically mentioning their existence brings on the unspeakable consequences, then I would say further: I find it perfectly possible to refer to and defend the right of freedom of speech by defending the expression of ideas with which I vehemently disagree. Hopefully Nazism is a safe enough bugbear to assume that every reasonable person disagrees with it. And yet I think Nazis should be able to speak, and at the same time that they should experience the consequences of that speech; and I feel safe in saying that no-one here will confuse my saying that with support or defence of Nazism. So why can't one of these things that can't be said similarly be named without supporting it?
To carry it even one step further, I agree that, if someone says that there's a high-pitched noise and I don't hear it, then it is rude to deny that they hear anything. But is it rude to ask them what it sounds like (for example, does it sound like something I can hear?), or if they can identify where it comes from, or if they hear it all the time? Maybe that falls under the category of 'demanding' evidence that pg says is rude, but it seems to me to be rather a gathering of evidence that I think is the sort of activity highly valued in this community.