About Twine
We're a thriving ecosystem of top-tier freelancers from around the world. Trusted by Fortune 500 companies and Silicon Valley startups, Twine is the go-to platform for mission-critical projects. Twine's experts get to participate in generating quality datasets at an unprecedented scale. With over half a million registered freelancers and a growing roster of 35,000+ companies, we've become the comprehensive solution for all things freelancing.
Our Mission
At Twine, we recognize that companies require a flexible approach to hiring expert freelancers for their most critical projects. Traditional freelance platforms often fall short when it comes to scalability and diversity. That's where we shine. Twine operates as a thriving global freelance network, with diverse experts across various fields, including marketing, engineering, and AI. Our core mission is to empower creators, whether they're businesses or individual freelancers, to thrive in their creative endeavours.
What is the project?
Are you a professional voice artist fluent in Japanese, Dutch, or French Canadian? We are looking for talented individuals to join a unique audio recording project for a client’s internal Text-to-Speech (TTS) research and development. Please note: This is an internal, non-broadcast, non-public project, with no paid or external TTS usage. Your voice will be used solely for AI model training.
Project Details:
- Audition Sample: Required upon application (script provided, approx. 150 words).
- Recording Requirements: Minimum of 3 hours of recorded audio.
- Script & File Management: Approved applicants will receive scripts totalling 3 hours. You’ll be responsible for maintaining natural pauses, correcting misreads, and submitting one .wav file per script following exact labelling instructions.
- Timeline: Once approved, you’ll have 4 business days to complete the recordings.
Audio Specifications:
- 24-bit, 48 kHz, mono
- 60dB noise floor (or better)
- Unprocessed, raw .wav files with minimal background noise, breaths, and clicks.
- Breaths do not need to be edited out but must be naturally minimized in performance.
- No post-production editing or plugins are allowed as they compromise audio quality.
Requirements
- Fluent in one of the following languages: Japanese, Dutch, or French Canadian
- Japanese speakers must not have previously recorded for our other TTS projects.
- French Canadian speakers specifically (not standard French).
- Age: 18+ (no upper age limit).
- Audio Specifications: You must be able to meet the high audio specifications detailed above.
Belgium
