AI evaluator - Audio transcription and evaluation
This experience involved utilizing OpenAI Whisper to transcribe and analyze video content, contributing to the detailed evaluation of dialogue flow and messaging clarity. Tasks included reviewing and refining transcripts for accuracy and clarity, as well as using transcriptions for further AI-assisted content evaluation. The workflow required a strong focus on structured reasoning, instruction accuracy, and data-driven evaluation methods. • Transcribed and evaluated large volumes of audio content using OpenAI Whisper. • Assessed and improved the logical consistency of AI-generated transcripts. • Collaborated in designing funnel systems for categorizing labeled data. • Applied prompt engineering techniques to optimize AI transcription outputs.