<![CDATA[Manual recording of the meeting minutes is often ineffective and has a high error rate. It makes access to needed information not fast. This research tries to solve these problems by designing and developing an end-to-end mobile application called Notulen.AI, which combines voice-to-text and Natural Language Processing (NLP) to automatically generate meeting minutes and summaries. The application was developed using the Flutter framework, with the Google Gemini API mainly used as the Automatic Speech Recognition (ASR) service and for NLP analysis. The research methodology consists of system design, module implementation, and testing to see performance and effectiveness. Testing on ASR modules gets good results with an average Word Error Rate (WER) of 6.75%. The NLP module also works well in extracting important information with ROUGE-1 scoring at 0.78 and F1-Score at 0.85. Effectiveness testing involving five respondents showed that this application could reduce minute-taking time by up to 70% and got a System Usability Scale (SUS) score of 85.5, which indicates high user acceptance. This research therefore proves that the integration of ASR and NLP on a mobile platform can be an efficient solution to enhance documentation accuracy in meetings.]]>
Copyrights © 2025
