This study introduces a framework for digitizing meeting content to support long-term storage and enhance semantic retrieval efficiency. The system integrates advanced speech processing technologies such as Automatic Speech Recognition (ASR), speaker diarization, and speech-to-text conversion. Whisper is applied to process conversational data in real-world environments, ensuring high accuracy even under noisy conditions and with low-resource languages like Vietnamese. A standout feature of the system is the speaker diarization module. This enables precise speaker identification even during overlapping speech and supports a fully automated end-to-end processing workflow without the need for manual clustering. The post-processed data is stored in structured formats within SQL database management systems, ensuring scalability and flexible integration. Additionally, the semantic retrieval tool allows content- and context-based search, surpassing the limitations of traditional keyword-matching approaches. This framework holds significant potential for widespread application in education, administration, business, and research.