I would like an option to precisely align the timing of the generated vocal performance.
Our director has requested the ability to quantize the timing of the rendered vocal sound to perfection. Although there is quantization available in the piano roll, we couldn’t find a feature to apply such quantization to the generated vocal audio.
Thank you in advance.
If you overwrite the default phoneme timing to that extent and eliminate preutterances it’s very likely the output will sound off-time and have poor understandability.
If that’s intentional, you’ll have to separate every lyric into single phoneme notes and set their durations one-by-one. There’s no simple way to do this because it’s typically an undesirable thing to do.
This example isn’t quantized, but shows how single-phoneme notes adhere to the note boundaries more strictly:
Source: Note and Phoneme Timing - Synthesizer V Studio Unofficial User Manual
To echo what claire has already said, when singing, the vowel is what aligns with the beat, not the consonants. There are exceptions, such as when there are liquid consonants such as /l/ or /w/.
That effectively means that consonants that are written at the start of a syllable are actually placed in time prior to the start of the note.
As claire stated, not aligning the vowel to the start of the syllable will actually give the wrong timing.