Conversational Speech Data Collection Project - Native Australian English
Budget: $25 – $50 USD
We are seeking freelancers for a real-world conversational speech data collection project in English. Participants must be native speakers of Australian English, able to work from anywhere in the world, and available to record in-person conversations with a partner.
The project focuses on capturing natural two-party conversations across general and medical domains to support advanced language research. All recordings must be conducted using a single microphone setup in the same physical location.
Key Requirements
- Native speaker of English (Australian en-AU)
- For the medical domain, a background in healthcare or life sciences is preferred (e.g., medical students, lab staff, nurses, or healthcare professionals)
- Must be able to pronounce medical terms clearly and naturally
- Must record with a partner in the same room using one device
- Conversations must sound natural and unscripted with no long pauses or overlapping speech
- Minimum 1 hour of recording (approx. 3 scripts) required to qualify for payment
Technical Details:
- File format: .wav
- Sample rate: 16kHz
- Audio channel: Mono
- Recording duration: Each conversation should be 5–20 minutes long
- Recording limits:
- Maximum of 2 hours per speaker
- Maximum of 4 hours per speaker pair
- Environment: Quiet room with clear sound and no background noise, echo, or distortion.
Payment Details:
- Rate: USD $38.00/hour
- Payments are calculated based on approved audio hours submitted to the client
- The monthly work cycle is based on completed recordings verified by the client
- Payments will be processed between mid and end of the following month after verification
Additional Information:
Payments will be managed through Freelancer's Milestone Payment System. A small placeholder milestone of $5 will be set up initially for each freelancer. Final payments will be adjusted and released based on actual hours worked, as reported by our client at the end of each cycle.
Rejection Criteria:
Recordings may be rejected if they do not meet project requirements, including but not limited to:
- Incorrect file format or sample rate.
- Stereo or edited audio files.
- Recordings made remotely (e.g., through Zoom or separate devices).
- Non-native accent, robotic or scripted delivery.
- Mispronunciation of medical terms or unnatural flow.
- Background noise, echo, or distortion affecting clarity.
Disclaimer
Participants must meet all technical and linguistic requirements to qualify for payment. Those who fail to meet the standards or use AI-generated or synthetic voices will be removed from the project and will not receive payment for their work.
All payments to participants will be processed securely through Freelancer.com.
The project focuses on capturing natural two-party conversations across general and medical domains to support advanced language research. All recordings must be conducted using a single microphone setup in the same physical location.
Key Requirements
- Native speaker of English (Australian en-AU)
- For the medical domain, a background in healthcare or life sciences is preferred (e.g., medical students, lab staff, nurses, or healthcare professionals)
- Must be able to pronounce medical terms clearly and naturally
- Must record with a partner in the same room using one device
- Conversations must sound natural and unscripted with no long pauses or overlapping speech
- Minimum 1 hour of recording (approx. 3 scripts) required to qualify for payment
Technical Details:
- File format: .wav
- Sample rate: 16kHz
- Audio channel: Mono
- Recording duration: Each conversation should be 5–20 minutes long
- Recording limits:
- Maximum of 2 hours per speaker
- Maximum of 4 hours per speaker pair
- Environment: Quiet room with clear sound and no background noise, echo, or distortion.
Payment Details:
- Rate: USD $38.00/hour
- Payments are calculated based on approved audio hours submitted to the client
- The monthly work cycle is based on completed recordings verified by the client
- Payments will be processed between mid and end of the following month after verification
Additional Information:
Payments will be managed through Freelancer's Milestone Payment System. A small placeholder milestone of $5 will be set up initially for each freelancer. Final payments will be adjusted and released based on actual hours worked, as reported by our client at the end of each cycle.
Rejection Criteria:
Recordings may be rejected if they do not meet project requirements, including but not limited to:
- Incorrect file format or sample rate.
- Stereo or edited audio files.
- Recordings made remotely (e.g., through Zoom or separate devices).
- Non-native accent, robotic or scripted delivery.
- Mispronunciation of medical terms or unnatural flow.
- Background noise, echo, or distortion affecting clarity.
Disclaimer
Participants must meet all technical and linguistic requirements to qualify for payment. Those who fail to meet the standards or use AI-generated or synthetic voices will be removed from the project and will not receive payment for their work.
All payments to participants will be processed securely through Freelancer.com.