Accelerating the delivery of high-quality software remains a significant challenge in the age of digital transformation. Software quality often depends on efficient requirements engineering (RE) processes and user input collection. Typically, software analysts use written notes, flip charts, pictures, and user reviews for gathering materials from RE sessions and user feedback. At so, these assets must to be converted into clear and valuable requirements for software development and testing. The manual nature of these tasks, however, often results in delays and impairs software quality attributes such as clarity, usability, and reliability. This paper introduces Requirements-Collector, a tool in automate requirements specification and user feedback analysis. The tool uses deep learning (DL) and machine learning (ML) techniques to accurately determine requirements from user feedback and audio recordings from RE session. These techniques are well suited for this purpose as they are particularly effective at text classification tasks. The role of software analysts could be significantly altered by the Requirements Collector since it saves repetitive tasks, enhances efficiency in communication, and takes us spare time for analytical work. In turn, this enhances the quality of software from its start. According to the study’s results, the tool correctly classifies requirements and input, presenting a significant step forward in automating the requirements engineering process.