$10–30/hr · micro1
You record and produce professional Thai audio content for AI training datasets, ensuring technical precision and linguistic authenticity.
What you would do
- Set up and operate professional recording equipment to capture clear, high-quality Thai speech content
- Edit audio files to meet project specifications including format, loudness, noise floor, and other technical requirements
- Review recordings for Thai pronunciation accuracy, naturalness, intonation, and technical defects
- Organize and deliver files following strict naming conventions and metadata standards for integration into training systems
- Respond to feedback and iterate on recordings to meet quality and specification standards
Who they want
- Native or near-native Thai fluency with strong listening skills for Thai language nuances and speech authenticity
- Minimum three years professional experience recording and editing speech or language audio content
- Access to and proficiency with professional-grade recording and editing equipment and software
- Strong written and verbal communication skills to coordinate with remote teams and clarify technical requirements
- Demonstrated attention to detail with proven ability to produce consistent, polished audio deliverables at scale
Main skills
What the interview asks about
1.Thai language recording quality
Native Thai ear is essential to distinguish authentic speech from delivery errors, unnatural pacing, or pronunciation that would confuse language models.
For example: “You receive feedback that one recording of 15 sentences sounds 'off' but the waveform looks clean. How would you diagnose and address the issue?”
2.Equipment setup and troubleshooting
Recording quality depends on setup: microphone placement, gain levels, room acoustics. You must get this right on the first take to avoid re-recording work.
For example: “Your first batch shows clipping on peaks but the waveform looks normal. What equipment or technique issue might be happening, and how would you correct it?”
3.Specification adherence and file organization
ML pipelines break if files don't match specs: bit depth, sample rate, naming format. One wrong file corrupts downstream processing.
For example: “The spec requires 44.1kHz mono, PCM WAV, with files named THAI_001_sentence.wav through THAI_100_sentence.wav. You've recorded everything but saved some as MP3 at 48kHz. How would you remediate this?”
4.Production pacing with quality maintenance
You're expected to hit production targets without quality slipping. You must identify when to push forward and when to slow down and re-record.
For example: “You're scheduled to record 20 sentences by end of day. After 12, you notice your delivery is getting tired and pacing is rushing. You have 3 hours left. What do you do?”
A task you may get
Conduct a practice recording session: set up your equipment, record 5-10 Thai sentences, edit one file to project specifications (demonstrate bit depth, loudness, file naming), and explain your quality checks for linguistic accuracy and technical compliance.
How to prepare
- Test your recording equipment setup and document baseline audio quality, identifying any consistent noise or technical issues to correct
- Record 2-3 sentences in Thai, listen for pronunciation and delivery authenticity, then re-record once to understand how you iterate on quality
- Review a sample project specification (format, loudness, naming convention) and practice organizing audio files to exact standards
- Schedule a practice session with your equipment, noting how many polished sentences you can record and edit in a realistic time window
The facts
- Pay
- $10–30/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
- Other
- Role type
- Generalist
- Posted
- 7/30/2026
- Places left
- 25
We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.