Geoffrey Kariuki

Key Skills

Software

Telus

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Top Label Types

Text Generation

Transcription

Freelancer Overview

I bring hands-on experience in data annotation and linguistic review, having worked as a freelancer on projects involving audio transcription, pronunciation validation, image tagging, and text-audio correction for platforms like TELUS International and Neevo DefinedCrowd. My work has included quality checks on audio files for native language, accent, and speech naturalness, as well as detailed transcription annotation with emotion and non-verbal cue tagging, and precise time-stamping of speech segments. With a strong background in data analysis and meticulous attention to detail developed through over 10 years in accounting and audit roles, I excel at maintaining high standards of accuracy and consistency in data labeling tasks. I am proficient with tools such as Microsoft Office Suite and have experience working with structured data and documentation systems, making me well-equipped to contribute to AI training data projects in domains like natural language processing and computer vision.

Labeling Experience

Project Sonic

TelusAudioText GenerationTranscription

Project sonic was a large scale data annotation and AI training initiative supporting the development and improvement of machine learning and NLP models.The project focused on producing high quality labeled dataset used to enhance model accuracy,relevance and reliability across language and content understanding tasks.Depending on the workstream the following annotation were involved,text and language annotation.audio annotation,pattern and image extraction,Removal of a person in a group,content evaluation.All tasks were performed using detailed annotation guidelines,annotation platforms and structured workflow.I handled high annotation volumes with tight turnaround times.The project followed rigorous quality control including multi layer quality assurance and performance metrics.The specific data labeling task performed were 2145.

Project sonic was a large scale data annotation and AI training initiative supporting the development and improvement of machine learning and NLP models.The project focused on producing high quality labeled dataset used to enhance model accuracy,relevance and reliability across language and content understanding tasks.Depending on the workstream the following annotation were involved,text and language annotation.audio annotation,pattern and image extraction,Removal of a person in a group,content evaluation.All tasks were performed using detailed annotation guidelines,annotation platforms and structured workflow.I handled high annotation volumes with tight turnaround times.The project followed rigorous quality control including multi layer quality assurance and performance metrics.The specific data labeling task performed were 2145.

2025 - 2025

Education

Work History