Hi, Chay.

Originally Posted by WaoBand
I really like the instrumentation and production in this very 70s disco sounding song - I can imagine Barry White singing it - but there's something about SynthV vocals that don't work for me. I suspect that it works a lot better in Japanese language songs than it does In English, but I always find with others SynthV vocals posts that there's no emphasis on syllables which should be emphasised as a real singer would emphasise them, plus not necessarily in this case, but I've found in others songs the final word or syllable at the end of lines so often sounds flat. Otherwise the song is great and really evokes that era.
Well, as long as the synth vocalist demos the song without being too distracting, it's a success.

Japanese is much different than English in a lot of ways that lend itself to being simpler to synthesize. There are less phonemes in Japanese, and for linguistic reasons, less phoneme combinations. It also lends itself to syllabic synthesis.

Emphasis comes from a number of sources, by primarily from the location of the word on the beat. It also takes into account syllable length and note pitch relative the prior pitch.

I'm not hearing the issues you are with emphasis - at least, not on this song. I suspect what you're hearing is the fact that it's synthetic.

I have heard issues with this in songs by non-native English speakers. But it's not a failing with SynthV, it's just that those writers don't hear the emphasis and stress of the words the same as an native English speaker. SynthV can't correct this, because it would typically be fixed by moving the position and duration of notes. A lot of this is subjective.

For example, I spent a fair amount of time on this song making sure that the phrasing sounded natural. Well, as natural as I could make something that's obviously artifice - words in a sung form. Plus, there were things that when sung by this particular voice didn't sound right, so I changed them to better match the singer.

But again - it's all subjective, and the voice is synthetic, and there's no getting around taht. So the uncanny valley is going to come poking through in places. That's why some people are doing their songs in SynthV, and then taking them to finalize the vocals. Suno simply has better voices than SynthV.

Thanks for your feedback, I appreciate it! smile

-- David Cuny
My virtual singer development blog

Vocal control, you say. Never heard of it. Is that some kind of ProTools thing?