I’m wondering how much I should trust acoustIDs to merge recordings. For example, those two recordings have the same acoustID: https://musicbrainz.org/recording/981a3c9b-bb93-43bf-a944-8e9ca04ca020 https://musicbrainz.org/recording/70749ed5-9e62-4989-8084-e79eb0f6914f and are recordings of the same work, is it evidence enough to merge them?
Merging classical Recordings: basis for sufficient confidence?
I wouldn’t https://www.amazon.com/gp/product/B0041W6VXI?tag=musicbrainz0d-20 gives “Lithuanian Chamber Orchestra and Aleksander Santi” for the same work. That said, there’s a chance one of the credits is wrong or both are pseudonyms for the same performers.
AcoustID is generally a good hint, but I wouldn’t merge based on that alone. I would merge based on that + same artists, and sometimes that + same label (if one of the two is in a compilation). Assuming similar durations and same work, obviously