Daniel Victor

about

I’m a multilingual linguistic QA and AI content evaluation specialist with over eight years of experience improving speech, image, video, and multilingual datasets. I work in English, Korean, German, and Spanish, helping teams evaluate content for accuracy, consistency, coherence, and linguistic quality. My experience includes high-volume speech labeling, audio quality assessment, video and image annotation in CVAT, and reviewing speech-to-text outputs for dropped content, hallucinated words, and unnatural pauses. I’ve applied annotation guidelines with 98% accuracy, corrected labeling errors, and created benchmark datasets for prosody detection and linguistic evaluation. I also train and supervise remote annotators, refine labeling protocols, and run QA calibration workshops to align teams on evaluation standards. With a master’s degree in linguistics and a background in phonetics and language technology, I bring careful analysis, clear feedback, and practical quality control to AI data projects.