We are seeking a fluent Japanese-speaking annotator for a human in the loop (HITL) data annotation role. Your primary role will be to conduct quality assurance and ensure data accuracy by auditing and refining automated speech-to-text transcriptions and brand categorization against actual video/audio content. This role demands native-level Japanese fluency to navigate complex language nuances, distinguish between similar brand names based on context, and accurately categorize mentions where automated tools might fail.

The successful candidate will be the final point of validation, ensuring the client receives clean, accurate data regarding brand mentions and sponsorship context.

• Video Verification: Audit automated text-to-speech transcriptions by watching broadcast video clips

• Brand Tagging & Sponsor Validation: Confirm the correct brand is accurately tagged.

• Type Classification: Verify if the classification of each brand mention was:

• Organic (spoken naturally by the talent/commentators) or
• Ad/Commercial: A scripted mention, paid placement, or commercial spot.

• Contextual Analysis: Proactively identify false positives where a word sounds like a brand name but isn't referring to the actual sponsor.

What We’re Looking For:

• Japanese Fluency: Native or near-native fluency in Japanese is mandatory. You must be able to understand rapid conversational speech, slang, and cultural context.
• Brand Awareness: Ability to recognize and distinguish between major global and prominent Japanese local brands.
• Attention to Detail: Ability to spot subtle differences between a literal word and a brand name.
• Technical Fit: Skilled in using Google Sheets for data entry, tracking, and organization.