"Is there even truth at all?"

I'd argue that Bayesian epistemology answers your question, to the extent that it is a coherent question.

To summarize, under Bayesian epistemology, you assign probabilities to statements. For example, you could assign 99% confidence to the statement, "the Sun is larger than the Earth." As you acquired evidence for that statement, you would adjust the probability upwards (perhaps to 99.9%). If you encountered evidence against that statement, you would shift the probability downwards again.

An important consequence of this system is that you NEVER assert absolute certainty. You always admit a slight possibility of error. Perhaps you're living in the Matrix, or perhaps all your life you've been missing some huge, important fact about reality. It's not likely, but it's possible.

In theory, a claim is true if (and only if) it accurately describes reality. In theory, there's an objectively "real" universe out there, and there exist statements that accurately describe said universe. In theory, you should be able to consider a claim, look at the universe, and objectively determine whether the claim reflects the universe. In theory, that's all there is to "truth."

In practice, it falls apart at the "look at the universe" step. In practice, you can't be certain that your view of the universe is unobstructed, or that you aren't completely misinterpreting a fundamental law of physics, or that you aren't simply misreading your instruments. You can't be certain that your memories are correct, or that a given scientific paper is mistake-free, or that you aren't speaking a slightly different language from everyone else because you learned mistaken definitions for some words.

In practice, the best we can do is be "pretty sure" that we have all the details right. We can be pretty sure that the Sun is bigger than the Earth, but never entirely certain. (And by "pretty sure," I mean "over 99% sure.")

And if that's the best we can do, then it has to be enough. We'll never get to 100% certainty, so probabilities around 99% will have to count.

tl;dr: "Truth" exists as an ideal to aspire to, and accumulated evidence is enough to call something "true" for practical intents and purposes.