OpenTrain AI

Compensation

Full Description

OpenTrain partners with leading AI teams to build and improve next-generation speech-recognition and language-learning systems. We’re hiring contractors for a high-precision French speech transcription project focused on word- and phoneme-level alignment. What you’ll do • Transcribe large batches of ~1-minute French audio clips into accurate word-level transcripts and ARPABET phoneme-level annotations using a secure web-based annotation tool • Identify and mark reading errors and disfluencies (e.g., repetitions, hesitations, false starts) with high accuracy • Work in rolling batches with peer-based QA review prior to final acceptance • Follow clear transcription and annotation guidelines to ensure consistency at scale Requirements (must-have) • Fluent French / Native level (written and spoken) • Expert-level ARPABET phonemic transcription skills • Proven ability to spot reading errors and disfluencies with high accuracy in spoken audio • Able to start within 1 week • Willing to pass a qualification quiz before beginning work • Able to commit 20+ hours/week at the published rate Preferred background • Phonetics / linguistics training (academic or professional) • Experience with ASR datasets or speech annotation pipelines • Experience with child-speech transcription (or similar challenging speech domains) Why this matters Your annotations will directly support the quality of advanced speech-recognition and language-learning models—work that impacts real-world products and research.