Citizen science, powered by artificial intelligence and machine learning, offers promising and positive avenues for biodiversity conservation. Our work presents a human-centric approach to biodiversity conservation and bioacoustic monitoring by classifying audio segments into broad categories (e.g., noise, animal calls) to improve annotation efficiency for endangered species identification. Hotspots like South Africa are an open but important problem because these species are bioindicators of the ecosystem’s health. Still, existing models primarily trained on common species often overlook them. Detailed annotations of audio data by volunteers is time-consuming, labor-intensive, and requires training. Importantly, annotators more often than not, must be able to label by sound alone. To tackle this challenge, data was collected in the wild across various sites over several months, generating vast and noisy datasets that require efficient processing. The sheer volume of data makes manual annotation impractical and presents a significant challenge. This highlights the need for an automated solution that appropriately integrates with annotators. We present a retrospective of our four-year project working on these issues. We demonstrate how competing demands for high quality data, annotator retention, and minimizing annotator effort all shaped the way our research was translated into practical innovations. Across all stages of our project, we highlight how our findings and innovations generalize to both other projects focused on acoustic data and other citizen science projects. We discuss some of the compromises made to balance these demands, and highlight the limitations of some of our decisions. This project emphasizes human interaction by automating labor-intensive tasks, allowing conservationists to focus on species identification and ecological analysis. The societal impact is profound—by streamlining the detection of endangered species, we contribute to global biodiversity monitoring and support conservation initiatives in regions under ecological threat.