Speech & Accent Evaluator – (Canadian) English- Location:Canada

About OpenTrain

OpenTrain aggregates data-labeling and AI-training jobs from many companies into one place so you can discover and apply quickly. Creating an OpenTrain account is free and applying takes only a few minutes.

This posting is for a contract project run through OpenTrain AI that focuses on improving spoken-language models in Canadian English.

About AI training and this work

AI training (also called data labeling or human feedback work) is the human side of building machine learning systems. People listen to, rate, and correct model outputs so speech systems become more natural, accurate, and safe.

This role involves listening to AI-generated audio, judging accent, fluency, and pronunciation, and giving clear, actionable feedback so the model can be iteratively improved.

The role

We are hiring Speech & Accent Evaluators who are based in Canada and use Canadian English daily. You will interact with an AI system in Canadian English by providing audio prompts and evaluating its spoken responses.

This is an entry-level, part-time contractor role (< 20 hrs/week) paid at USD $25 per hour. Labeling work is audio-based and uses an external/other labeling platform provided by the project.

• Employment type: Contractor, part-time.
• Pay: $25 USD per hour (hourly, PAY_PER_HOUR).
• Weekly time commitment: Less than 20 hours/week.

What you'll do day-to-day

Your primary task is to evaluate AI-produced speech in Canadian English and provide structured feedback to improve accent, pronunciation, and fluency.

You will also conduct interview-style assessments of candidates and/or simulated interactions where you act as an AI interviewer evaluating linguistic expertise and evaluation ability.

• Listen to AI-generated audio and assign evaluation ratings (EVALUATION_RATING) for accent, fluency, and intelligibility.
• Provide concise, structured comments describing errors, unnatural phrasing, or pronunciation issues and how to fix them.
• Record and submit audio prompts as required by the project workflow.
• Run iterative evaluation cycles: assess output, suggest improvements, request/produce new model responses until target quality (e.g., 7/7) is met.

How the interview-style assessment works

Part of this project requires you to act as an AI interviewer. You'll follow a professional-but-conversational tone to keep candidates engaged while collecting the information needed to judge their skills.

You will use a structured script during assessments: greeting and process overview, questions about experience, phonetic/feature probes, iterative-improvement scenarios, and closing questions.

• Greet the candidate and explain the assessment goals and format.
• Ask about length and context of their use of Canadian English and probe accent-specific features and intonation patterns.
• Present iterative scenarios: explain how you'd guide multiple model attempts to reach a perfect 7/7 for accent and fluency, and what you'd focus on each iteration.
• Close by asking if the candidate has questions, thanking them, and describing next steps.

Requirements and skills

You must be physically located in Canada and have at least 2 years of experience using Canadian English in daily communication (personal or professional). This location requirement ensures familiarity with local pronunciation and accent variation.

The role is entry level; a background in linguistics, phonetics, speech analysis, or language teaching is a plus but not required. Strong attention to detail and the ability to produce clear, structured feedback are essential.

• Must be based in Canada (required).
• At least 2 years using Canadian English daily (personal or professional).
• Entry-level experience acceptable; domain study in linguistics/phonetics is beneficial but optional.
• Strong written communication, ability to deliver structured feedback, and keen ear for pronunciation differences.

Evaluation criteria we use

Your submitted evaluations and interview assessments will be judged on several key dimensions to ensure high-quality contributions to model improvement.

These criteria explain what we expect when you rate and provide feedback on audio.

• Linguistic Accuracy: Demonstrates knowledge of Canadian English accent features and pronunciation.
• Speech Evaluation Skills: Able to identify accent, fluency, and pronunciation issues in AI audio.
• Iterative Improvement Mindset: Can propose and guide multiple focused iterations to refine outputs.
• Communication Clarity: Provides structured, actionable feedback that modelers can use.

How to apply and next steps

To apply, create a free OpenTrain account and submit your application for this posting. Shortlisted candidates will be contacted with instructions for a brief evaluation exercise and onboarding into the project's labeling platform (OTHER).

After onboarding you'll receive assignments, submit audio prompts and ratings through the project interface, and be paid as a contractor at the stated hourly rate.

• Create an OpenTrain account and apply through the job listing.
• Be prepared for a short assessment that tests your Canadian English familiarity and evaluation approach.
• Onboarding includes using the project's labeling software (listed as OTHER) and following the provided workflow.