Unfortunately, "quatrevingt" won't sound exactly like "quatre" followed by "vingt" (in "quatrevingt", you will make the sound for the "e" very clear while mutin

Unfortunately, "quatrevingt" won't sound exactly like "quatre" followed by "vingt" (in "quatrevingt", you will make the sound for the "e" very clear while muting it in "quatre"). Of course, you could just use more developed algorithms to connect words but then why not just use a voice synthesizer base only on recorded syllables?