Levin K.

Key Skills

Software

Other

Appen

Top Subject Matter

Localized Swahili/Nairobi Sheng linguistic audio transcription and AI data evaluation

Linguistic quality evaluation

transcription verification

Top Data Types

Audio

Text

Document

Top Task Types

Transcription

Freelancer Overview

Freelance Transcriptionist & AI Data Evaluator (Remote). Brings 3+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Other and Appen. Education includes Bachelor’s Degree, KCA University and ALX AI Career Essentials Certification, ALX Africa (2026). AI-training focus includes data types such as Audio and labeling workflows including Transcription, Evaluation, and Rating.

Labeling Experience

Freelance Language & Data Specialist (CrowdGen / RWS / Appen)

AppenAudio

Freelance language and data specialist conducting linguistic quality assessments and verification tasks for global AI crowdsourcing platforms. Reviews and edits machine-generated transcriptions and performs audio-to-text alignment to improve speech recognition model training data quality. Applies native-level Swahili expertise and strong English proficiency to ensure grammatical and auditory comprehension in dataset outputs.• Linguistic quality assessments, translation verifications, and audio-to-text alignments for crowdsourcing datasets.• Review and editing of machine-generated transcriptions to support speech recognition model accuracy.• Completion of RWS/CrowdGen language and transcription assessments for Swahili and English.• Ensures dataset reliability through rigorous verification and elite comprehension checks.

2025 - Present

Freelance Transcriptionist & AI Data Evaluator (Remote)

OtherAudioTranscription

Freelance speech annotator and verbatim transcriptionist for large-scale, localized linguistic AI projects, including handling multi-speaker accents and Nairobi Sheng. Performs precise speech-to-text annotation and transcription while following client styling guides and strict data privacy requirements. Delivers consistently high-quality outputs, averaging 99.2% quality scores across submitted datasets.• High-fidelity audio transcription and speech annotation for localized language datasets.• Verbatim transcription of multi-speaker audio with timestamps and structured text formatting.• Review and alignment to project-specific guidelines for accuracy and confidentiality.• Quality assurance against client benchmarks with reported 99.2% average quality score.

2025 - Present

Education

Work History