I'm the anon who said in a q reseach thread earlier today I would have speech therapist listen to audio and compare. This expert did not think they were same person:

Nasality is different between the 2 voices and and that is hard to fake/alter.

The idiolect is different.

Suprasegmentals of voice (pitch and reverb) are also different. However, these can be altered/faked easily.

Fricative consonant length does compare pretty good, but could compare to a large segment of population.