Speech AI Evaluation Specialist

About the role

This role is tailored for Thai residents in Bangkok who wish to perform remote freelance evaluations of Vietnamese speech data from home. You will engage in short voice-based interactions with AI systems using provided scenarios and prompts to test system behavior. The position requires a commitment of more than ten hours per week with flexible scheduling, allowing you to start immediately and work hour-by-hour. You will assess AI responses based on relevance, factual accuracy, speech clarity, and conversational quality using structured criteria. By rating these interactions and documenting observations, you directly contribute to improving the safety and reliability of AI-generated Vietnamese content. This opportunity is ideal for students, recent graduates, stay-at-home parents, and gig workers in Thailand seeking supplemental income. Your work will help shape the development of contemporary AI models by ensuring outputs meet strict quality and guideline standards.

Key facts

What you'll do

• Execute evaluation tasks for by conducting Vietnamese voice interactions.
• Access the official apply page and strictly adhere to the scope and instructions provided for the evaluation workflow.
• Conduct short voice conversations with AI models, setting dialogue contexts and guiding conversation flow according to project instructions.
• Apply clear rating criteria to assess AI output for clarity, relevance, factual accuracy, and speech quality during each session.
• Document every observation using structured feedback formats to ensure consistency and traceability in evaluation records.
• Track productivity metrics diligently to maintain high accuracy levels and verify that responses align strictly with project guidelines.
• Coordinate with international teams regarding evaluation timelines, deliverables, and any adjustments to project requirements.
• Set up appropriate dialogue contexts and testing scenarios to ensure comprehensive coverage of AI interaction capabilities.
• Update status logs promptly to reflect completed evaluation batches and communicate any issues or anomalies discovered.
• Follow data evaluation and quality assurance processes to ensure that collected data remains valid and useful for model improvement.
• Provide constructive feedback based on objective criteria to help refine AI behavior and improve future Vietnamese content interactions.
• Maintain strict compliance with project rules, ensuring that all work meets the defined standards for clarity, accuracy, and relevance.

Requirements

• Must possess native-level fluency in Vietnamese to accurately evaluate linguistic nuances, tone, and context during speech interactions.
• Must demonstrate fluent or advanced proficiency in English, corresponding to levels B2 through C2 on the Common European Framework of Reference for Languages.
• Must be a resident of Thailand, specifically based in Bangkok, to fulfill remote work expectations and time coordination.
• Must commit to a minimum of ten hours per week on a flexible schedule that allows you to choose your working hours.
• Must be able to start immediately and remain available for an assignment with a duration to be confirmed.
• Must accept a compensation rate of 18 USD per hour for all evaluation work completed.
• Must have the capability to perform remote evaluations from a home setup using a personal computer and stable internet connection.
• Must understand and apply data annotation and labeling practices to ensure evaluation outputs are precise and usable.
• Must be familiar with data collection and preprocessing methods to support effective participation in AI evaluation cycles.
• Must have a clear understanding of data quality assurance processes to identify inconsistencies or errors in AI responses.
• Must be a student, recent graduate, stay-at-home parent, gig worker, or other professional in Thailand seeking flexible remote employment.
• Must follow detailed project guidelines rigorously to ensure consistency across all evaluation tasks and international team expectations.

Nice to have

• Prior experience in AI and data capabilities, including familiarity with machine learning tasks and evaluation workflows.
• Experience in data collection, preprocessing, and handling speech datasets for training or testing purposes.

Practical notes

• This is a hourly contract role compensated at 18 USD per hour with flexible scheduling options.
• The position is fully remote and conducted from home, requiring only a stable internet connection and a computer.
• Candidates must be available to work from Thailand, with special consideration for residents of Bangkok.
• Visa arrangements are not applicable due to the remote nature of the contract and local residency requirements.
• There are no specific deadlines for application submission, but early application is encouraged due to the ongoing nature of the evaluation cycle.
• Hours are flexible but must exceed ten per week to ensure consistent contribution and productivity.
• Coordination with international teams may require adjustment of working hours to align with project timelines.