Cantonese Transcriptionist for AI Training

Contribute professional Cantonese and English transcription expertise to projects that support the training and improvement of advanced AI systems. This remote contract role focuses on producing accurate, high-quality linguistic data from real-world audio, with particular attention to dialect, context, speaker nuance, and transcription quality.

• Transcribe and annotate Cantonese and English audio recordings with a high degree of accuracy and consistency.
• Review and refine automatically generated transcripts to ensure they follow linguistic and project guidelines.
• Identify and correct language nuances, speaker accents, dialectal variation, and contextual cues in audio samples.
• Produce clean, time-stamped transcripts that meet required formatting and quality standards.
• Work with project leads to resolve ambiguities and clarify unclear audio content.
• Provide linguistic insights that help improve AI training data and model performance.
• Complete and submit assigned work according to established project timelines and milestones.

• Native proficiency in Cantonese, including strong understanding of regional dialectal variations.
• Demonstrated professional audio transcription experience, particularly with Cantonese and English content.
• Strong attention to detail, accuracy, and linguistic sensitivity.
• Proficiency with transcription tools and audio playback software.
• Ability to type in both traditional and simplified Chinese.
• Strong written and verbal communication skills, with an emphasis on clarity and completeness.
• Experience reviewing or annotating linguistic data for technology, media, or research projects is a plus.
• Ability to work independently, manage assignments, and consistently deliver high-quality results in a remote environment.
• Prior AI experience is not required.

• Remote contract engagement.
• Work is project-based and involves contributing expert transcription and linguistic input to AI training initiatives.

Pay ranges from $20 to $35 per hour.

Candidates can apply online using an available account or supported sign-in method. Proceeding with the application requires acknowledgment of the applicable candidate terms and privacy policy.