$30–65/hr · micro1
A Vietnamese language expert who evaluates audio fluency and authenticity for AI training purposes.
What you would do
- Listen to audio clips in Vietnamese and score speaker fluency and linguistic naturalness
- Assess pronunciation precision, tonal accuracy in tonal language, and sentence-level intonation
- Write detailed feedback in English explaining linguistic factors underlying each evaluation
- Provide clear justifications for native vs. non-native distinctions based on observable speech features
- Collaborate with the project team to clarify evaluation criteria and resolve edge cases
Who they want
- Native-level fluency in Vietnamese with deep command of spoken language and regional variations
- Minimum B2 proficiency in English for written and verbal communication with clarity and precision
- Background in voice specialization, linguistics, phonetics, music engineering, or language teaching
- Experience with audio analysis, linguistic assessment, or transcription preferred
- Meticulous focus on linguistic nuances and skill in describing subtle phonetic and prosodic distinctions
Main skills
What the interview asks about
1.Tonal accuracy and tone-bearing unit assessment
Vietnamese tones are contrastive; incorrect tone placement changes word meaning and signals non-nativeness immediately.
For example: “You hear a speaker say a word with the rising tone when the falling tone was called for in the script. How would you describe this error in English for someone who doesn't speak Vietnamese?”
2.Distinguishing native and model-generated speech
Machines often produce phonetically correct but prosodically odd speech; detecting this protects training data authenticity.
For example: “You listen to two clips of the same sentence. One sounds natural but has a subtle pronunciation shift; the other is perfectly clear but slightly rhythmically stiff. How would you evaluate each?”
3.Regional dialect and register variation
Vietnamese has regional dialects and formal/informal registers; evaluating speech requires distinguishing non-native from regional authenticity.
For example: “A speaker uses Northern Vietnamese phonology but applies Southern intonation patterns inconsistently. Is this non-native speech, code-mixing, or a regional blend? How would you document it?”
4.Phonetic precision and segmental clarity
Small articulation errors accumulate; training models on imprecise speech undermines pronunciation accuracy in learned systems.
For example: “A speaker consistently palatalizes final consonants slightly, which is close to standard but noticeably off. Write feedback explaining what's happening and whether it's critical or minor.”
5.Structured English documentation of linguistic findings
Your English explanation must be clear enough that non-linguists and non-Vietnamese speakers can understand why a score was given.
For example: “You need to explain to a non-speaker why a clip failed fluency evaluation. Write a 2-3 sentence justification focusing on observable features, not just 'sounds foreign.'”
A task you may get
Listen to 2-3 Vietnamese audio clips and provide individual assessments covering pronunciation, intonation, fluency, and authenticity, with written justification in English for each score.
How to prepare
- Review the International Phonetic Alphabet notation for Vietnamese consonants and tones; practice describing tonal and articulation differences precisely.
- Listen to native Vietnamese speech across regions and registers to internalize what naturalness sounds like in varied contexts.
- Practice writing technical linguistic observations (tone patterns, vowel quality, stress placement) in clear English for non-linguist readers.
- Study common non-native error patterns in Vietnamese speech to build a mental checklist of diagnostic features.
The facts
- Pay
- $30–65/hr
- Open to
- Bangladesh, Hong Kong, India, Indonesia, Japan, Kazakhstan, Kyrgyzstan, Malaysia, Pakistan, Philippines, Singapore, Sri Lanka, Taiwan, Thailand, Uzbekistan, Vietnam, Austria, Belarus, Belgium, Denmark, France, Germany, Greece, Italy, Netherlands, Portugal, Russia, Spain, Switzerland, United Kingdom, Argentina, Brazil, Chile, Colombia, Mexico, Peru, Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Tunisia, United Arab Emirates, United States, Canada, Nigeria, Kenya, South Africa, Ghana, Ethiopia
- Field
- Language Audio
- Role type
- Specialist
- Posted
- 8/13/2026
- Places left
- 100
We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.