Classification of Spontaneous and Scripted Speech for Multilingual Audio