Modern inventions have made it possible to hear how our great grand parents spoke. Will this influence how the language and dialects change?

*Probably* not.

Language change is influenced by several things, in both a conservative and a innovative direction. Input from older versions of the languages demonstrably has an effect in holding back language change — or at least, in promoting use of the older version’s features in parallel.

  • Outright reversing language change doesn’t happen that often, and needs special circumstances—like with Icelandic and flæmeli (small population, universal literacy). Conservative influence however has a lower threshold for success.

But the success of conservative influence is incidental to it being older. The real reason why any conservative input would be successful is that it is being held up as prestigious. This is what happens with standard literary versions of languages: they happen to be more conservative than spoken variants of languages, but they influence language change because they are held up as prestigious, particularly in education.

So for old recordings to influence language change, it is not enough that they become available. They would need to be actively promoted in mass education as models to be emulated. In western culture at least that seems unlikely.

One area where recordings have much more of an impact is language revival efforts. In that context, knowing what your ancestors’ native accent was like is very important, though it may not be enough for you to shake off your modern accent…

Is it possible to make a language out of only one type of word (noun, verb, adjective etc)?

Logan R. Kearsley has written a comprehensive answer on one angle. I will throw a hint on another angle: if you have enough Noun Incorporation (linguistics)  and polysynthesis at a language, you’re going to end up with languages where what European languages treat as nouns or adjectives usually end up as affixes—so what look like words are mostly verbs. In fact, from time to time you do hear people claiming that some such languages (almost always Amerindian languages) don’t have nouns, though my recollection is that the claim is marginal.

So you can in principle have languages with just verbs. However in practice you do have affixes that will tend to signify arguments rather than predicates—so you’ve really just pushed the noun/verb distinction down into morphemes rather than words.

Oh, I see Logan has also written on the converse, whether a language can have just nouns: Is it possible to make a language with just nouns and adjectives?