The answer is C. It is clear that computers are not "learning" to recognize spontaneous speech, because the developers are using just sample materials
↧
.jpg)






