AI Audio Annotator / Data Labeler, Wehive (via DingTalk Platform)
Provided high-precision audio annotation and transcription to support AI training datasets in both Bahasa and English. Completed complex labeling tasks to create ground truth for downstream speech and language model improvements. Maintained quality by following project guidelines and linguistic nuances while flagging edge cases for engineering review. • Phonetic transcription and transcription labeling for multilingual audio • Speaker identification labeling for training data • Sentiment tagging for labeled audio segments • QC-driven review to sustain high accuracy and identify edge cases