tech note ‐ music is music, phonetics is phonetics (also why you don't need legacy plugins) - stakira/OpenUtau GitHub Wiki

Traditionally, music, lyrics, and phonetics are in the same layer in UTAU. In order to get better results, users have to split notes to convert notes from hiragana lyrics to diphone or n-phone (such as CV, VC, VCV) aliases. This destroys the original lyric or the musical meaning: quarter notes, eighth notes, etc. And often users need to convert diphone notes back to musical notes, then convert to diphone notes again based on another set of rules. A swarm of plugins, and maybe half of the most used ones, are made for this. Besides, some technical challenges are entirely caused by this conversion step, such as splitting pitch curves, splitting envelopes, or even merging them, which sometimes are missions impossible.

So let's face it: music is music, phonetics is phonetics. A central concept of editing in OpenUtau is phonemizing. It separates the music & lyric layer from the phonetics layer. Notes are kept in the original form, with proper note length and actual lyrics, carrying the pitch curves, vibratos, etc. Users can pretty much reuse the notes when switching between phonetic systems or voicebanks. The splitting merging problem doesn't exist anymore. Phonemizer works in real-time, making editing interactive and easy.

It is often asked whether legacy plugins are supported. The true answer is you won't need the majority of them. Also if you consider that many of the editing experiences have been reworked (which learned a lot from some most used plugins) in OpenUtau, you won't need most of the legacy plugins. Of course, there is always the possibility to do better and more. OpenUtau offers APIs (though some are still in plan) for plugin development, which works with the OpenUtau data model better (meaning no need to convert to UST then back, which is lossy), and integrates with the user experience better.