French ARPABET Phoneme Transcription Specialist

About OpenTrain

OpenTrain is the #1 platform for building careers in AI training and data labeling. Contributors find projects, build a profile, and apply quickly to work that helps shape how modern AI systems understand language and speech.

We connect skilled people with hands-on annotation work across industries. This project is part of that mission—supporting speech recognition and language-learning models with high-quality human transcriptions.

About AI training work

AI training (data labeling) is the human side of building artificial intelligence. Annotators create and verify examples—like phoneme-level transcripts—that models learn from. These roles are usually 100% remote, flexible, and accessible to people with language expertise or domain knowledge.

This role focuses on phonetics and ASR training: your fine-grained judgments about pronunciation, disfluency, and reading accuracy will directly improve speech recognition and language-learning systems.

Role overview

We are hiring native-level French speakers to transcribe and align French audio at both word and phoneme (ARPABET) levels. Work is performed in a web-based annotation tool and delivered in rolling batches.

This is a contractor, part-time role paid hourly. You must be available for 20+ hours per week and able to begin within one week of acceptance.

• Data type: French audio (long clips, ~1–2 minutes each)
• Tasks: word-level alignment and phoneme-level (ARPABET) transcription and validation
• Pay: $18–20 USD per hour (hourly rate; paid for approved work)
• Commitment: 20+ hours per week, ongoing
• Location: Remote — global
• English: Conversational
• Employment types: Contractor, Part-time
• Labeling software: Web-based tool (Other)

What you'll do

Your annotations will be used to train next-generation ASR and language-learning models. Work arrives in batches and goes through peer QA before final acceptance.

• Listen to 1–2 minute French audio clips and produce precise word- and phoneme-level (ARPABET) transcriptions
• Create accurate word-level alignments and map pronunciations to ARPABET phonemes
• Detect and label reading errors: substitutions, omissions, insertions, and misread words
• Identify and label disfluencies: filled pauses, repetitions, false starts, and self-corrections
• Follow project guidelines and incorporate peer QA feedback to maintain high consistency and accuracy
• Deliver work in rolling batches and meet throughput targets while preserving quality

Requirements

To be successful you must combine native French proficiency with practical experience in phoneme-level transcription and careful attention to spoken detail.

• Native-level French with strong sensitivity to fine-grained pronunciation and reading accuracy
• Intermediate to expert proficiency with ARPABET phonemic transcription for French speech
• Proven ability to identify reading errors (misreads, substitutions, omissions, insertions)
• Proven ability to identify disfluencies (filled pauses, repetitions, false starts, self-corrections)
• Prior experience in at least one: phonetics/linguistics, ASR dataset transcription, phoneme-level labeling, or child-speech transcription
• Able to produce consistent word-level alignment and phoneme-level transcription with high attention to detail
• Availability to start within one week of acceptance and commit 20+ hours/week
• Must pass a qualification quiz before receiving production batches
• Comfortable working in a web-based annotation tool and participating in peer QA loops

How work is structured & paid

Work is delivered in rolling batches through a web annotation platform. All submissions go through peer QA; only approved work is paid. High accuracy and steady throughput increase your approved hours and earnings.

You will take a qualification quiz before starting production. After onboarding you will receive batches and feedback from peer reviewers to help you reach project standards.

• Qualification quiz required before production
• Assignments delivered in rolling batches; peer QA review before final acceptance
• Paid hourly at $18–20 USD per hour for approved work
• Higher accuracy and consistent throughput improve overall earnings

How to apply

If you meet the requirements and can start within a week, apply to join this project. Be prepared to show examples of past phoneme-level work or related transcription/linguistics experience during the qualification process.

After application you will be asked to complete a short qualification quiz; passing the quiz unlocks production batches and access to the web annotation tool.

• Prepare samples or a summary of relevant experience in phonetics, ASR transcription, or child-speech work
• Expect to complete a qualification quiz and brief onboarding
• Commit to 20+ hours/week and a start date within one week of acceptance