Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Image

Text

Top Task Types

Audio Recording

Freelancer Overview

With extensive hands-on experience in audio data transcription and labeling for AI training, I have developed a rigorous, detail-oriented approach to ensuring data quality. My work involves meticulously transcribing speech according to strict style guides, with a particular focus on accurately applying a complex set of tags (e.g., [fp], (()), <ga>, <ms>) to denote speech characteristics, non-verbal sounds, and audio quality issues. I am highly proficient in adhering to language-specific rules, such as spacing conventions for Chinese, Japanese, and Korean (CJK) text, and consistently applying guidelines for handling unintelligible speech, cross-talk, and speaker diarization. What sets me apart is my demonstrated ability to quickly internalize complex labeling protocols and consistently apply them with high accuracy, as evidenced by my proficiency in tasks ranging from marking filled pauses and truncations to making critical save-state decisions (Good/Discard). My skills are not merely technical; they are rooted in a strong analytical ability to discern subtle audio events and a commitment to producing clean, reliable training data that is essential for building robust speech recognition and natural language processing models. This combination of precision, adaptability, and a deep understanding of data integrity makes me a valuable contributor to any AI data pipeline.

Labeling Experience

Mandarin Audio Transcriber – Simplified Chinese

LabelboxAudioTranslation LocalizationAudio Recording

​Project Distribution: Audio Data Transcription and Labeling Workflow​​ ​​1. Project Initiation & Setup (5%)​​ Review and internalize comprehensive style guide, tagging conventions, and language-specific rules. Familiarize with the annotation platform/tool interface and submission protocols. Establish quality control checkpoints and understand save-state criteria (Good/Discard). ​​2. Core Audio Processing & Transcription (60%)​​ ​​Task Allocation & Listening (10%):​​ Receive and queue audio files of varying durations. Conduct initial careful listening to assess content, audio quality, and identify key characteristics. ​​Primary Transcription (35%):​​ Execute verbatim transcription of intelligible speech in the target language (e.g., Simplified Chinese). Apply strict formatting rules: lowercase letters, no punctuation, correct CJK word spacing, and normalization of numbers/acronyms.

​Project Distribution: Audio Data Transcription and Labeling Workflow​​ ​​1. Project Initiation & Setup (5%)​​ Review and internalize comprehensive style guide, tagging conventions, and language-specific rules. Familiarize with the annotation platform/tool interface and submission protocols. Establish quality control checkpoints and understand save-state criteria (Good/Discard). ​​2. Core Audio Processing & Transcription (60%)​​ ​​Task Allocation & Listening (10%):​​ Receive and queue audio files of varying durations. Conduct initial careful listening to assess content, audio quality, and identify key characteristics. ​​Primary Transcription (35%):​​ Execute verbatim transcription of intelligible speech in the target language (e.g., Simplified Chinese). Apply strict formatting rules: lowercase letters, no punctuation, correct CJK word spacing, and normalization of numbers/acronyms.

2025 - 2025

Education

Work History