Chaolian Li

Key Skills

Software

Appen

Label Studio

OneForma

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Video

Top Task Types

Audio Recording

Translation Localization

Freelancer Overview

As an expert in data labeling with 3+ years of experience, I specialize in multi-modal annotation (audio, video, text) across industries like ICT, healthcare, and autonomous driving. I’ve led teams in delivering 10M+ high-quality labels using tools like Appen, Label Studio, and custom platforms, with a focus on NLP intent classification, computer vision object detection, and quality control frameworks that ensure annotation accuracy ≥98%. My background in linguistics and computer science enables me to design efficient labeling workflows and adapt to emerging AI training data needs.

Labeling Experience

Data Annotator

AppenVideoAudio Recording

This project focused on video/audio data labeling in the ICT industry. The scope included transcribing and annotating 5,000+ video-embedded audio recordings related to telecommunications equipment operations and network troubleshooting scenarios. Specific tasks involved speech-to-text transcription with a word error rate threshold of ≤2%, intent classification for identifying technical support requests, and entity tagging for equipment names and error codes. The project size covered 200 hours of video audio, and quality measures included double annotation for 20% of samples (inter-annotator agreement ≥95%) and regular audits using Appen’s built-in quality control modules.

This project focused on video/audio data labeling in the ICT industry. The scope included transcribing and annotating 5,000+ video-embedded audio recordings related to telecommunications equipment operations and network troubleshooting scenarios. Specific tasks involved speech-to-text transcription with a word error rate threshold of ≤2%, intent classification for identifying technical support requests, and entity tagging for equipment names and error codes. The project size covered 200 hours of video audio, and quality measures included double annotation for 20% of samples (inter-annotator agreement ≥95%) and regular audits using Appen’s built-in quality control modules.

2021

Education

Work History