Le Dang Khoa

Key Skills

Software

Remotasks

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Text

Video

Top Task Types

Action Recognition

Audio Recording

Computer Programming Coding

Data Collection

Text Summarization

Freelancer Overview

I’m a mid-level AI training data specialist with 1+ years’ experience in multilingual annotation and dataset creation. I’ve led projects spanning code annotation, audio transcription, text summarization, and conversational AI prompt generation—delivering high-quality, noise-filtered recordings and metadata in English, Vietnamese, and Japanese. Comfortable with tools like Labelbox and CVAT, I ensure each asset meets rigorous QA standards, consistently achieving ≥98% accuracy on spot checks. My expertise lies in crafting authentic conversational datasets and annotating complex technical material. I excel at designing role-play scripts, timestamping and intent-tagging utterances, and structuring JSON metadata for seamless ML ingestion. By combining strong linguistic skills with a methodical QA process, I help clients accelerate model development and improve AI understanding across diverse domains.

Labeling Experience

Multilingual Conversational Audio Data Labeling Specialist

Scale AIAudioTranslation LocalizationPrompt Response Writing SFT

Built a 5,000-utterance, multi-scenario conversational audio dataset (10 scripts) to train an AI speech model. Handled scriptwriting and recording in English–Vietnamese–Japanese, intent & metadata tagging, and noise-filtered QA with ≥98% pass rate. Delivered structured WAV files, transcripts, and JSON metadata for ML ingestion.

Built a 5,000-utterance, multi-scenario conversational audio dataset (10 scripts) to train an AI speech model. Handled scriptwriting and recording in English–Vietnamese–Japanese, intent & metadata tagging, and noise-filtered QA with ≥98% pass rate. Delivered structured WAV files, transcripts, and JSON metadata for ML ingestion.

2024 - 2024

Education

Work History