Thanks, Andy, and Wikidatans,
Where free, CC-4 MIT OCW-centric World University & School, having donated itself to Wikidata in 2015 for co-development, would like to head with languages without a written form, and also with the languages of people who don't write, conceptually, is in the direction of 10-person Google group video Hangouts into Youtube, so that phonemes/sounds could become part of your query (and also into sounds transcribed to graphemes/written forms). And WUaS would like our eventual matriculated students - linguists and anthropologists, for example - to go out to the field and generate these further.
The main languages' wiki subject page at WUaS -
https://wiki.worlduniversityandschool.org/wiki/Languages - plans to create wiki subject pages for open teaching and learning for all 8,475 entries in languages / 7,099 living languages (in Ethnologue), and also develop an Universal Translator between these, too, perhaps partly developing from all 113,021 code points (in conjunction, conceptually, with GNMT/Google Translate).
Is there a way to query for Wikidata phonemes in a parallel way to your query for Wikidata graphemes? Thanks for this query.