姚夷颢

Key Skills

Software

Other

Top Subject Matter

General AI/NLP

General AI/Computer Vision

General AI/Speech Recognition

Top Data Types

Text

Image

Audio

Document

Top Task Types

Classification

Bounding Box

Transcription

Freelancer Overview

AI Data Annotator – Text Classification and Review. Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other. Education includes Junior College Degree, Communication University of China. AI-training focus includes data types such as Text, Image, and Audio and labeling workflows including Classification, Bounding Box, and Transcription.

Labeling Experience

Audio Transcription & Annotation Specialist

OtherAudioTranscription

I transcribed and labeled audio clips, including speaker identification and noise filtering tasks. I verified transcript accuracy by cross-checking against source audio and correcting errors. My work contributed to producing reliable transcripts for speech recognition model training. • Annotated complex audio with homophones and context-dependent content. • Created a guide for common transcription challenges. • Assisted the team in improving labeling consistency. • Logged errors and maintained records using spreadsheets.

Present

AI Image Annotator – Bounding Box & Landmark Annotation

OtherImageBounding Box

I performed bounding box and landmark annotation on image datasets to support AI model training. Each image was labeled according to strict Intersection over Union (IOU) standards. My precise work ensured the dataset met the high accuracy requirements for computer vision projects. • Reviewed and corrected team members' bounding box annotations. • Maintained consistency by following project-specific guidelines. • Identified and corrected common annotation errors. • Documented process improvements for workflow optimization.

Present

AI Data Annotator – Text Classification and Review

OtherTextClassification

I labeled over 10,000 text samples for intent classification and sentiment analysis for AI training purposes. I carefully followed project guidelines to maintain consistency and accuracy across all samples. My work directly supported the creation of high-quality training datasets for large language models. • Achieved a 98%+ accuracy rate on annotated samples. • Conducted secondary reviews, reducing the overall project error rate by 15%. • Maintained thorough documentation of edge cases and challenging samples. • Used spreadsheets to track progress and errors.

Present

Education

Work History