French ARPABET Phoneme Transcription Specialist (Native French)

About OpenTrain

OpenTrain is a centralized job board for data-labeling and AI-training roles. We aggregate openings from many AI teams and labeling platforms so you can find audio, transcription, and annotation jobs in one place. Creating an account and applying takes only a few minutes.

About AI training and why it matters

AI training (also called data labeling or annotation) is the human work that teaches models to understand speech, text, and images. For speech recognition, skilled annotators create precise transcripts and phoneme alignments that models use to learn accurate, natural-sounding recognition.

This project supports next-generation French ASR and language-learning systems by producing high-quality phoneme-level and word-aligned transcripts.

The role

You will convert long French audio clips (typically 1–2 minutes) into precise word-level and ARPABET phoneme-level transcripts using a web-based annotation tool. Work is provided in rolling batches; each batch you complete will go through peer QA before final acceptance.

Location: Global — this is fully remote. Required English level: conversational. Employment type: contractor, part-time. Time requirement: 20+ hours per week on an ongoing basis.

• Data type: audio (French speech phonetics and ASR training).
• Label types: audio recording and transcription (word-level alignment and phoneme-level).
• Tooling: web-based annotation platform with peer QA loops.

Compensation & schedule

Pay rate: $18–$20 per hour (per the published hourly range). Work is delivered in batches and annotations are approved through a QA process; higher throughput and consistent accuracy lead to better earnings over time.

You must be available for at least 20 hours per week and able to start within one week of acceptance. This is ongoing contract work with flexible hours as batches are available.

What you'll do day to day

Your primary task is to produce accurate, consistent transcripts at both the word and phoneme (ARPABET) level and to flag reading errors and disfluencies. You will follow detailed labeling guidelines and respond to peer-QA feedback to improve quality.

• Listen to 1–2 minute French audio clips and produce word-level transcripts.
• Transcribe phonemes using ARPABET conventions for French speech.
• Mark reading errors such as substitutions, omissions, insertions, and misread words.
• Label disfluencies: filled pauses, repetitions, false starts, and self-corrections.
• Submit work via the web-based tool and respond to peer QA comments as needed.

Requirements

You must meet every essential qualification below to be eligible for this project. All items come from the project brief and are required for selection and continued work.

• Native-level French: able to judge fine-grained pronunciation and reading accuracy.
• Intermediate to expert proficiency with ARPABET phonemic transcription for French speech.
• Proven ability to identify and label reading errors (misreads, substitutions, omissions, insertions).
• Proven ability to identify and label disfluencies (filled pauses, repetitions, false starts, self-corrections).
• Prior experience with at least one of: phonetics/linguistics work, 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.
• Must be able to start within one week of acceptance.
• Must pass a qualification quiz before receiving production batches.
• Availability to commit 20+ hours per week on an ongoing basis.
• Comfortable working in a web-based annotation tool and following a peer-QA feedback loop.

Who should apply and next steps

Apply if you are a native French speaker with phonetics or ASR transcription experience who can commit to the schedule and quality standards above. This role is best for people who enjoy careful, detail-oriented audio work and collaborative QA processes.

Next steps: create an OpenTrain account, submit your application, complete the qualification quiz if invited, and be prepared to begin within one week if accepted.