TADA: Speech generation through text-acoustic synchronization