$30–65/hr · micro1 · Part time
A native Japanese speaker with strong English proficiency who records spoken Japanese sentences and vehicle sound effects for machine learning audio datasets.
What you would do
- Record yourself reading Japanese sentences with natural pronunciation and precise enunciation in a quiet office or from a stationary vehicle
- Create authentic background audio including seatbelt clicks, door sounds, car indicator signals, and other environmental noises while maintaining safety protocols
- Follow detailed linguistic guidelines covering sentence content, pacing, intonation, and delivery to ensure consistency across recordings
- Submit high-quality audio files meeting technical specifications for format, bitrate, sample rate, and metadata
- Provide written feedback on your recordings and collaborate with project coordinators on revisions or clarifications
Who they want
- Native-level proficiency in Japanese with clear, natural enunciation and excellent pronunciation of phonemes and intonation patterns
- Minimum B2-level English proficiency for written and verbal communication with project coordinators
- Ability to follow precise technical and linguistic guidelines consistently across multiple recording sessions
- Self-directed work habits and ability to manage personal time and deliverables autonomously
- Attention to detail regarding audio quality, pronunciation accuracy, background noise, and data integrity
Main skills
What the interview asks about
1.Japanese pronunciation authenticity
Your native speech patterns and clear articulation directly affect whether the AI model learns accurate Japanese phonology. Flaws in pronunciation or unnatural pacing reduce dataset quality.
For example: “Recording 'r', 'l', 'ts', and 'dz' sounds: how do you ensure natural, consistent pronunciation across multiple takes and what indicates a problem?”
2.Sound effect recording safety
You must create realistic vehicle sounds while keeping the car stationary. You need to understand which sounds are safe to make and how to capture them without causing distraction or hazard.
For example: “Recording door, seatbelt click, and indicator sounds safely from a parked car: describe your approach while maintaining quality and safety.”
3.Technical audio guidelines
Your recordings must meet specific technical requirements to integrate with the training pipeline. Misunderstanding file format or quality specs results in rejected submissions.
For example: “16 kHz WAV with metadata (speaker ID, take number, content type): how do you verify your software settings are correct before submission?”
4.Guideline compliance consistency
Linguistic consistency across recordings ensures the AI model encounters predictable patterns. Drift in pronunciation, pacing, or delivery reduces the signal-to-noise ratio in training data.
For example: “Fifth session: your pacing has quickened and you've varied accents to avoid monotony. Is this a problem and what should you do?”
5.English-Japanese collaboration
Project coordinators communicate primarily in English, but your expertise is in Japanese. You need to bridge this gap clearly to resolve issues and ensure deliverables meet expectations.
For example: “A coordinator says an 'o' vowel sounds too closed. You understand acoustically but the English is imprecise. How do you seek clarification?”
6.Attention to background noise
Audio from a quiet office or stationary car should be clean. Unexpected noise (HVAC, traffic, notifications) can degrade recordings and require re-recording, which affects your hourly output.
For example: “Air conditioning kicks on during recording at 10 AM. Noise is audible but subsides minutes later. Do you continue, re-record, or reschedule?”
A task you may get
Record three Japanese sentences with clear pronunciation and a one-minute audio file with five vehicle/environmental sounds. Submit with notes on conditions and challenges.
How to prepare
- Test recording setup: microphone, software settings (sample rate, bit depth, format), and audio levels for technical compliance
- Practice Japanese at various speeds and intonation to identify what sounds natural and where you naturally drift
- Research vehicle sounds (doors, seatbelt, indicators) production and test capturing them safely with your equipment
- Assess your B2-C1 English skills for clarity and precision communicating with technical coordinators
The facts
- Pay
- $30–65/hr
- Hours
- Part time
- Where
- Remote
- 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
- Generalist
- Posted
- 8/13/2026
- Places left
- 80
We wrote this page from the public micro1 listing. It may be out of date, so read the full posting before you apply.