Multilingual AI Linguistic Testing -- 2

I’m finalising an AI-powered platform and need sharp-eared, detail-oriented linguists to put every language feature through its paces. Your mission is to chat with the model, listen to its speech output, inspect translations, and flag anything that sounds odd, unnatural, or simply wrong. Native-level proficiency in at least one of the target languages is essential so that slang, register, and subtle pronunciation issues don’t slip past us.

Target languages
• Chinese
• Japanese
• Korean
• English
• Spanish
• German
• Russian
• Swedish
• Polish
• Italian
• Dutch
• Romanian
• Arabic

What you’ll be testing
• LLM text responses for relevance, coherence, and tone
• TTS (Text-to-Speech) for clarity, natural rhythm, and correct pronunciation
• Forward and back translations for accuracy and cultural fit

Deliverables
1. A short screen-capture or audio snippet (if TTS is involved) showing the issue
2. A concise written explanation that pinpoints the problem and suggests a native-sounding fix
3. Completion of a feedback form for each test session so we can track recurring patterns

Acceptance criteria
• Every reported issue must be reproducible with clear steps or a timestamped clip
• Feedback should be written in English plus the target language when relevant
• Suggestions must be actionable (e.g., “stress the second syllable,” “replace with colloquial synonym”)

We work in iterative sprints; expect quick turnarounds and prompt follow-up questions from our engineering team. If you’re ready to use your linguistic instincts to make AI sound truly native, I’d love to see a brief note on which language(s) you’ll cover and any prior testing or localisation experience you can bring to the table.