Most data streaming systems collect and process real-time data from various sources. However, many decisions require both streaming and historical data to obtain a comprehensive view. Entity resolution (ER) determines whether two records refer to the same real-world entity in the absence of a shared identifier. Therefore, it is crucial to integrate stream data and stored data together through entity resolution and to ensure the accessibility and usability of data from different sources. Existing ER methods often fail to support incremental stream processing while efficiently handling complex data like text. Given the strength of embedding techniques in capturing semantic and syntactic information, we aim to adapt embedding-based ER to streaming data for integrating incoming and existing records. We propose a dynamic graph embedding suitable for stream processing to perform entity resolution within relational tables.