Divya H.

Key Skills

Software

Appen

Other

Top Subject Matter

Kannada speech data transcription and quality review

LLM response evaluation and quality assurance

Kannada audio data collection for speech model training

Top Data Types

Text

Audio

Top Task Types

Transcription

Audio Recording

Freelancer Overview

Transcription Specialist & Reviewer – Appen (CrowdGen, Jigglypuff Project). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Appen and Other. Education includes Bachelor of Engineering, K S School of Engineering and Management (2025). AI-training focus includes data types such as Transcription, Text, and Audio and labeling workflows including Transcription, Evaluation, and Rating.

Labeling Experience

Transcription Specialist & Reviewer – Appen (CrowdGen, Jigglypuff Project)

AppenTranscription

Reviewed and validated Kannada transcription outputs using detailed language and formatting guidelines. Flagged quality issues and provided feedback to maintain dataset accuracy and annotation consistency. Served as a reliability reviewer based on consistent quality during production work. • Produced accurate transcriptions following project rules • Corrected and validated transcription outputs • Checked formatting, language, and quality consistency • Provided structured feedback aligned with standards

2026 - Present

Audio Data Contributor (Kannada) – Silencio

OtherAudioAudio Recording

Recorded Kannada voice samples to contribute native language data for AI speech-training datasets. Followed recording instructions, pronunciation standards, and audio-quality requirements. Helped improve Kannada speech-recognition model training data. • Recorded Kannada voice samples for training sets • Met pronunciation and recording requirements • Ensured audio quality and formatting compliance • Contributed native Kannada data for speech recognition

2026 - 2026

LLM Evaluation Contributor – Appen (CrowdGen, Hazel Project)

AppenText

Evaluated AI-generated responses for factual accuracy, relevance, clarity, tone, safety, and guideline adherence. Identified hallucinations, unsupported claims, and instruction-following failures. Assigned structured ratings and wrote rationales to support model quality improvement. • Used evaluation rubrics to assess response quality • Detected hallucinations and factual inconsistencies • Scored responses and documented rationales • Ensured consistency across multi-task evaluations

2025 - 2026

Audio Data Contributor (Kannada) – Appen (CrowdGen, Moogerah Project)

AppenAudioAudio Recording

Contributed Kannada speech data for AI language and voice-model training by completing recording and contribution tasks. Followed project guidelines and quality standards for dataset collection. Supported creation of diverse, high-quality Kannada audio inputs. • Collected Kannada speech recordings per instructions • Followed guideline and quality requirements • Contributed data supporting voice-model training • Ensured consistency with dataset collection standards

2025 - 2025

Education

Work History