The primary product of the annotation process, these datasets are enriched with labels or annotations (e.g., bounding boxes for images, transcriptions for audio, or tags for text). Labelled datasets are the foundation for training, validating, and testing AI/ML models, tailored to the client’s specific use case.