This paper presents the development of an interactive tool for automatic speech transcription in Spanish and keyword detection, leveraging OpenAI’s Whisper model. The proposed system is designed to handle long and unstructured audio files by implementing a modular architecture that supports intelligent segmentation based on silence detection and customizable time intervals. Transcriptions are generated with precise timestamps and presented through a user-friendly web interface, allowing users to perform contextual searches for specific terms defined prior to execution. The front-end was built using Streamlit, while audio preprocessing and segmentation tasks were handled by the Pydub library, ensuring accessibility and ease of deployment. Experimental results demonstrate a strong positive correlation between audio duration and processing time \((R^{2} = 0.85)\) , as well as a competitive average Word Error Rate (WER) of \(8.12\%\) , even under challenging acoustic conditions. These outcomes confirm the robustness and practical applicability of the tool in real-world environments. The system provides a lightweight and adaptable solution for automated audio analysis, with potential applications in education, digital accessibility, media monitoring, forensic investigations, and legal transcription workflows.