English (New Zealand) Audio Generalist Evaluator

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We help people start and grow durable freelance careers teaching AI by bringing specialized projects, profile tools, and application workflows together in one place.

OpenTrain AI is the hiring and contracting organization for this role. Creating an OpenTrain account is free and lets you apply, manage work, and build a portfolio of AI training projects.

About AI training work

AI training (also called data labeling or annotation) is the human side of building intelligent systems. Contributors transcribe, annotate, and evaluate audio to help models learn accurate speech recognition, language understanding, and culturally aware outputs.

This work is accessible, in-demand, and often flexible: projects range from short tasks to ongoing part-time contracts and let you shape how state-of-the-art systems behave. Your contributions directly improve model performance for real users.

The role — what you'll be paid and where you must be located

OpenTrain AI is hiring an English (New Zealand) Audio Generalist Evaluator to transcribe, analyze, and evaluate New Zealand English audio for research and benchmark readiness.

• Pay: USD 50 per hour (PAY_PER_HOUR).
• Time: 20+ hours per week (part-time contractor).
• Employment type: Contractor, part-time.
• Location: Applicants must be located in Singapore (country eligibility: SG).
• Data & tasks: Audio work focused on TRANSCRIPTION and EVALUATION_RATING.

What you'll do day-to-day

• Listen to English (New Zealand) audio and video and produce accurate, well-formatted transcripts that follow provided style and formatting rules.
• Identify and capture tone, intent, register, regional expressions, Maori loanwords, and other NZ English nuances.
• Develop or refine evaluation rubrics and grading guidelines for general consumer audio contexts.
• Grade and test model outputs for accuracy, completeness, fluency, and instructional clarity against established criteria.
• Support QA cycles: resolve ambiguities, document edge cases, and provide feedback to improve task design before benchmark release.

Requirements

• Native or near-native fluency in English (New Zealand), spoken and written — required.
• Ability to transcribe spoken New Zealand English accurately, including varied accents and regional vocabulary.
• Strong writing, editing, and critical thinking skills.
• Familiarity with New Zealand English vocabulary, contemporary usage, and Maori influences in speech.
• Comfort evaluating model outputs against rubrics and quality standards.
• Preferred: prior experience with transcription, annotation, localization, evaluation, or research workflows.

Who should apply

This role is a strong fit for linguistically attuned English speakers in Singapore who can reliably transcribe New Zealand English and judge model outputs against clear quality criteria.

Apply if you enjoy careful listening, clear written work, and shaping how conversational AI understands regional English varieties.

How to apply and next steps

Create an OpenTrain account (free), complete your profile, and submit your application for this role. OpenTrain AI will review applications and contact qualified candidates with next steps.

Contracts are managed by OpenTrain AI; pay and scheduling details are established at onboarding. If selected, you'll receive task-specific guidelines and the tools needed to begin work.