ipa-dict VS dev

According to this UK data set, "sa" is same as in massage, can't find samples for "au" or short "u", but same "u" in long form is for example in adjutant and "asume, and they don't sound too long for me. Then it is just sliding from one vowel to other.

I am not aware of any. However, I have found this to be a good source for IPA data https://github.com/open-dict-data/ipa-dict. Based on that data I use the following function to convert text to IPA:

There is a list on github, just tap into that file to retrieve te pronounciation.

Credit to the open-dict-data (https://github.com/open-dict-data/ipa-dict) project for their extension dictionary of IPA spellings of English words.

https://github.com/open-dict-data/ipa-dict/tree/master/data (random word lists)

Does the language you’re thinking of have an inventory on https://phoible.org/?

I found this a while back. Does this help? https://phoible.org

Non-trivial problem, actually. As other people say rolling your own from a database like Phoible might be your best bet. There are some good summary statistics for phonemes in Matthew Gordon's 2016 'Phonological Typology' text, but I don't think he does detailed correlation tables.

I think in general voiceless stops and fricatives are more common than voiced ones. Source being, of course, PHOIBLE.

You may want to give Phoible a look for typological questions about frequency of occurrence for sounds.

You can use https://phoible.org/. In the segment part, there are percentages of use of phonemes of the recorded languages. On the other hand, I don't know if it's representative enough

I guess I’m looking for something like this: PHOIBLE but with audio

Here you go. You just need to download the full Phoible CSV (e.g. via the Github repository) and install dplyr/tidyr in R.