Language Data Annotator & Transcription Evaluator
As a Language Data Annotator and Transcription Evaluator, I evaluated and corrected LLM-generated transcriptions for conversational accuracy, coherence, and speaker intent. My role required applying structured feedback, detailed annotation taxonomies, and strict guidelines to ensure high dataset quality. I contributed to RLHF projects and provided expert human annotations to improve model performance. • Reviewed and compared multiple model-generated transcripts based on predefined rubrics. • Provided corrections for recognition errors, semantic inaccuracies, and formatting issues. • Applied complex labeling taxonomies and identified tone shifts and speaker overlaps. • Delivered structured written feedback to enhance AI conversational understanding.