Andromeda Delta - Audio Transcription Adjudicator (India/Philippines)
RemoteFreelance / independent contractor
Core
Review and resolve discrepancies in word-level transcription and timestamp annotations for US English speech data.
Role type
Adjudicator (QA)
Builds
High-quality speech datasets for training AI models
Domain
Speech technology / Audio data annotation
Deliverable
production ML models
Required skills
US English listening comprehension, adjudication judgment, guideline application, error pattern identification
Preferred skills
Prior transcription/annotation QA experience
Technologies
Audio playback tools
Responsibilities
Review annotations flagged by first-pass annotators, listen to audio clips and make final determinations on word-level segment boundaries, apply project style guide consistently, provide clear rationale for adjudication decisions, identify patterns in annotator errors
