Speech representation is one of the most important tools which is used by writers in writing narrative texts and specifically stories. Although direct and indirect speech, which are two speech representation categories, are familiar names in Persian language, speech representation has been neglected in Persian language. This article investigates and compares different speech representation categories used in Persian stories written before translation movement in Qajar era and stories written after this movement and assesses the frequency, functional and linguistic characteristics of each category. For this purpose, the Semino and Short’s model of speech representation (2004) which is based on Leech and Short’s model (1990) is used. The research data is elicited from 7 stories written before translation movement and other 7 stories written after this movement in equal volumes. Finally, we conclude that (free) direct speech is the most frequent category in the stories written before and after translation movement and this is because of the nature of this category in producing the effects of dramatization and vividness. Also, the high frequency of (free) direct speech in recent stories can be related to the key role of characters’ individuality in modern novels. Moreover, the use of (free) direct speech and narrator’s representation of speech acts as the speech representation categories have decreased during the time but the use of indirect speech and specifically free indirect speech have increased.