AI Voice Quality Evaluator

About the Role

We are looking for native or highly proficient Vietnamese speakers to evaluate AI-generated voice recordings and provide structured feedback on voice quality, expressiveness, and similarity.

As an AI Voice Quality Evaluator, you will listen to pairs of audio samples and compare them based on detailed evaluation criteria. Your feedback will help assess and improve the quality, naturalness, expressiveness, and voice similarity of AI-generated speech. This is a language and audio evaluation role. No technical or programming background is required.

Key Responsibilities

• Listen to and compare pairs of AI-generated audio samples using headphones.
• Evaluate audio samples based on predefined quality and evaluation guidelines.
• Assess basic voice quality, including pronunciation, prosody, intonation, stability, audio quality, and audio-text consistency.
• Evaluate expressiveness, including tone, emotion, intonation, conversational style, and character style.
• Evaluate voice similarity against a reference recording, including voice/timbre, speaking style, and accent similarity where applicable.
• Assign comparative scores based on the provided evaluation scale.
• Select appropriate reasons explaining the evaluation result.
• Provide concise and specific written comments describing key strengths, weaknesses, or noticeable issues.
• Follow project guidelines consistently and maintain a high level of attention to detail.

Requirements

• Native or near-native proficiency in Vietnamese language.
• Strong listening and language comprehension skills.
• Good sensitivity to pronunciation, intonation, tone, speaking style, and emotional expression.
• Able to identify subtle differences between two audio recordings.
• Strong attention to detail and ability to make consistent comparative judgments.
• Able to follow detailed evaluation guidelines accurately.
• Good written communication skills in English and/or the required project language.
• Comfortable working with audio evaluation tasks using headphones.
• Previous experience in AI evaluation, linguistic annotation, data annotation, localization, transcription, voice/audio evaluation, language QA, or similar work is a plus, but not mandatory.

Preferred Qualifications

• Background in linguistics, language studies, translation, interpretation, or related fields.
• Experience in speech, voice, dubbing, podcasting, broadcasting, or voice acting.
• Previous experience evaluating AI-generated speech, TTS, voice cloning, or conversational AI.
• Experience with multilingual or cross-cultural language evaluation.

Working Conditions

• Remote / project-based work.
• Flexible working arrangements depending on project requirements.
• Candidates will be required to use headphones during audio evaluation.
• Candidates must be able to complete repetitive listening and comparison tasks while maintaining consistent evaluation quality.