AI Data Labeler & Evaluator (OpenTrain AI / Freelance)
Labeled and evaluated English-language audio samples for naturalness, fluency, and benchmark adherence. Rated and annotated speech and transcription content while maintaining consistency across batches under JSON schema specifications. Collaborated asynchronously with global review teams to support AI model training and quality workflows. • Audio evaluation and quality/rating of AI-generated speech • Speech and transcription annotation under JSON schema • Naturalness and fluency assessment against English benchmarks • Async review collaboration to meet delivery timelines