Audi oSegmentation Trainer - Contract
The project involves Italian-language speech data labeling, focused on high-accuracy audio segmentation and verbatim transcription. Tasks consist of listening to short audio recordings and producing clean, standardized transcripts that strictly reflect the spoken content, following detailed general and locale-specific guidelines. The scope includes audio segmentation, transcription, and listening-based verification, such as identifying unclear or unusable segments and flagging portions that cannot be confidently understood using predefined options. Work is performed within a dedicated annotation platform and follows a consistent workflow designed to ensure reliability and repeatability. Project size is ongoing and long-term, with a steady daily workload. Quality is ensured through strict adherence to transcription and segmentation rules, consistency checks, careful judgment in cases of imperfect audio, and compliance with all provided guidelines and certifications when task-specific