Conversational Text-to-Speech (CTTS) aims to accurately express an utterance with the appropriate style within a conversational setting, which has attracted more attention nowadays. While recognizing the significance of the CTTS task, the prior studies have not thoroughly investigated the speech emphasis expression, which is essential for expressing the underlying intention and attitude in human-machine interaction scenarios, due to the scarcity of conversational emphasis datasets and the difficulty of context understanding. In this paper, we propose a novel Emphasis Rendering scheme for CTTS model, termed ER-CTTS, that includes two main components: 1) we take into account textual and acoustic contexts simultaneously, with both global and local semantic modeling to comprehensively understand the conversation context; 2) we deeply integrate multi-modal and multi-scale context to learn the influence of context on the emphasis expression of the current utterance. At last, the inferred emphasis feature is fed to the neural speech synthesizer to gen- erate the conversational speech. To address the data scarcity, we create emphasis intensity annotations on the existing conversational dataset (DailyTalk). Both ob- jective and subjective evaluations suggest that our model outperforms the baseline models in emphasis rendering within a conversational setting. The code and audio samples can be found at https://github.com/CodeStoreTTS/ER-CTTS.