The paper discusses the development and enhancements of ELAN, a professional framework for multimodality research, particularly focusing on time accuracy and flexibility. ELAN is designed to meet the stringent requirements of various linguistic sub-fields, such as gesture studies, sign language research, and multimodal interaction studies. The authors highlight the importance of precise timing in these fields, where even a 20 ms shift can lead to significant errors. To address these challenges, ELAN has been updated to integrate with Windows DirectX libraries, improving performance, accuracy, and stability. The tool now supports multiple media formats, including MPEG-2, and allows for the synchronization of up to four video streams. Additionally, ELAN offers advanced search capabilities, import/export modules for various annotation formats, and enhanced productivity features like semi-automatic segmentation and tier tokenizing. The paper concludes by outlining future plans for further enhancements, including structured search on multiple files, integration with lexicon components, and the incorporation of time series data from movement tracking devices.The paper discusses the development and enhancements of ELAN, a professional framework for multimodality research, particularly focusing on time accuracy and flexibility. ELAN is designed to meet the stringent requirements of various linguistic sub-fields, such as gesture studies, sign language research, and multimodal interaction studies. The authors highlight the importance of precise timing in these fields, where even a 20 ms shift can lead to significant errors. To address these challenges, ELAN has been updated to integrate with Windows DirectX libraries, improving performance, accuracy, and stability. The tool now supports multiple media formats, including MPEG-2, and allows for the synchronization of up to four video streams. Additionally, ELAN offers advanced search capabilities, import/export modules for various annotation formats, and enhanced productivity features like semi-automatic segmentation and tier tokenizing. The paper concludes by outlining future plans for further enhancements, including structured search on multiple files, integration with lexicon components, and the incorporation of time series data from movement tracking devices.