About the Role
The role involves recording spoken Basque in various contexts and evaluating audio data to improve machine learning systems' understanding of the language.
Responsibilities
- Deliver natural-sounding voice recordings in Basque across diverse scenarios
- Follow scripts accurately while maintaining authentic regional intonation
- Identify and flag pronunciation inconsistencies in audio datasets
- Provide linguistic insights on dialectal variations in Basque speech
- Review machine-generated transcriptions for accuracy
- Rate the quality of synthesized Basque speech samples
- Adapt vocal tone to match specified emotional or situational contexts
- Follow data privacy protocols when handling voice materials
- Meet deadlines for audio submission batches
- Collaborate with remote team members on language-specific challenges
- Report technical issues related to recording tools
- Maintain consistent audio quality across sessions
- Follow annotation guidelines for speech data
- Contribute to improving AI model performance through feedback
- Stay updated on project-specific requirements and workflows
Nice to Have
- Formal training in voice acting or linguistics
- Previous work with speech technology projects
- Experience in language teaching or translation
- Knowledge of phonetics or phonology
- Background in Basque cultural contexts
Compensation
Competitive hourly rate
Work Arrangement
Remote
Team
Distributed team focused on AI training and language data
Project Duration
Initial engagement is project-based with potential for extension based on performance and demand
Time Commitment
Approximately 10–15 hours per week, with flexible scheduling
Technology Requirements
- Computer running Windows, macOS, or Linux
- Stable internet connection
- External microphone recommended
Data Privacy
- All participants must comply with confidentiality agreements
- Recordings will be used solely for AI training purposes
Performance Evaluation
- Quality of submissions is monitored regularly
- Feedback provided to support ongoing improvement
Not applicable
