Voice Over Artist: AI Voice Data Training




Get Paid to Record for a Voice Data Project
Remote freelance opportunity to record scripted speech from home.

1 hour guaranteed, up to 10 hours total.
Recordings are used for internal AI research only – no public or commercial use.

Project Type: AI voice data collection (internal research & development only)
Recording Time: 1 hour guaranteed, up to 10 hours total
Pay: $120–$400 USD per hour (varies by language & location)
Location: Remote / work from home
Experience Required: Previous voice acting experience is preferred. Professional audio quality required.
Usage: No voice cloning. No public-facing AI. Internal testing only.

What You’ll Do
You’ll record scripted speech in your most fluent language to help improve conversational AI systems. 

  • Record approximately 9,000 words per finished hour
  • Deliver studio-quality audio from a quiet environment
  • Complete recordings within 2 days of assignment

This is project-based voice over work facilitated by Voices. These are not internal Voices positions.


Languages Needed

English, French (France), Spanish (South America), Portuguese (Brazil), German, Italian, Arabic, Hindi, Bengali, Vietnamese, Thai, Indonesian, Tagalog

You must be native or fully fluent in the language you apply for.


How Your Voice Will Be Used
Your recordings will be used strictly for internal AI research and system evaluation.

  • Used to prompt speech-enabled AI systems
  • Used to assess accuracy, safety, and response quality
  • Not sold, broadcast, or publicly distributed
  • Not used for commercial or public-facing AI
  • Not used for voice cloning, synthesis, or digital voice replicas

    Your voice will not be cloned, generated, or used in synthetic speech. All contributors sign an NDA and usage release. Data is stored securely and handled in compliance with privacy standards.

Audio Requirements
You must have access to a professional-quality recording setup.

  • WAV (.wav) format
  • 44.1 kHz or higher (48 kHz preferred)
  • 16-bit or higher (24-bit preferred)
  • Noise floor below -60 dBFS
  • Quiet, echo-free recording space
  • Raw, unprocessed audio only (no effects, noise reduction, or compression)

Compensation
Pay per Hour: $120-$400 USD per hour (varies by language & location)
Total Pay: $120-$400 guaranteed, $1,200 – $4,000 if hired for all 10 hours

  • 1 hour guaranteed
  • Up to 10 hours total
  • Payment issued via Tipalti (verified bank account required for Tipalti)

Apply
To apply, you will:

  • Submit a short audio sample (under 1 minute) to confirm recording quality
  • Complete a quick demographics survey

    AI-generated submissions are strictly prohibited. Audio samples are reviewed manually and with detection software. Submitting AI content or multiple auditions may result in removal from this and future projects without compensation.

Submit your audition at the following links:

We’ll review all submissions and contact selected candidates with the next steps!

Continued Opportunities

Even if you’re not selected this time, applying for this project sets you up for recurring work in your language as new matching Voice Data opportunities become available, and you join Voices’s roster of remote Voice Data contributors.

Voices is committed to collecting AI data ethically, with full speaker consent and fair compensation. Click here to learn more.