Improve chunk_text algorithm #739
Description
Description & Motivation
I (Jordan) have integrated the algorithm chunk_text (lightly adapted from https://github.com/fakerybakery/txtsplit) into EveryVoice synthesis, as a project during my work term. However, there are a few issues with the algorithm itself I haven't yet worked out, at least for our purposes. See the sub-issues...
Pitch
No response
Alternatives
No response
Additional context
No response