Voice Actor: CX Agent Voice Cloning (Spanish - Peninsular)

What you would do

• Record high-quality voice samples across conversational, narrative, and instructional scripts
• Deliver clear and expressive speech with strong control over emotional tone and pacing
• Maintain consistent Castilian accent and pronunciation across multiple recording sessions
• Execute recording guidelines for audio quality, environment, microphone setup, and file format
• Execute multiple recordings with varied emotional intensity and stylistic emphasis as instructed

Who they want

• Native Castilian/Peninsular Spanish speaker based in Spain
• Clear, natural, and expressive voice quality (prior voice acting or broadcast experience is a bonus)
• Access to professional or near-professional recording setup with quality microphone and quiet environment
• Proficiency with vocal inflection, pronunciation clarity, and expressive range across various conversational contexts
• Availability for approximately 4-hour recording session with possibility of second similar-length session

Main skills

What the interview asks about

1. 1.Emotional vocal delivery variety

   TTS models must learn to synthesize speech with appropriate emotional color for different contexts; your ability to shift tone while maintaining authenticity trains that nuance.

   For example: “Read this customer-service greeting five times: once very warm and welcoming, once calm and professional, once energetic and upbeat. Describe what changes in your delivery each time.”

2. 2.Castilian accent consistency

   AI learns regional dialect from training data; your consistent pronunciation of theta, theta sound, and regional inflection must be reliable across hours of recording.

   For example: “How do you ensure your pronunciation of 'gracias' versus 'cebolla' remains authentic and consistent over many takes without slipping into neutral Spanish?”

3. 3.Recording technical setup and quality

   Microphone technique, room acoustics, and file format directly affect AI training; you must manage these variables to produce clean, usable data.

   For example: “Your recording setup includes a USB microphone in a small home office. Walk through how you'd prepare the space and position the mic to minimize echo and background noise.”

4. 4.Script interpretation with natural pacing

   Verbatim script reading sounds robotic; AI learns naturalness from your phrasing choices, where you breathe, and how you stress words - that's what makes synthesis convincing.

   For example: “You have a 30-second script of a customer service response. Explain where you'd add micro-pauses, which words you'd emphasize, and how you'd vary your pacing.”

5. 5.Voice cloning consent and understanding

   This role explicitly involves voice cloning for AI; your comfort with that technology and clear understanding of how your voice will be used is essential.

   For example: “Explain in your own words what voice cloning means for your voice, how it will be used in this AI agent, and why your recordings won't be used elsewhere.”

A task you may get

Record a 60-second customer service scenario in three different emotional tones - professional-friendly, calm-reassuring, and energetic-helpful. Deliver each take with distinct vocal choices while maintaining clear Castilian pronunciation.

How to prepare

• Test and optimize your home recording setup: microphone placement, room acoustics, and audio software settings
• Practice delivering the same script multiple times with distinct emotional tones while keeping your accent consistent
• Research TTS systems and how voice samples are used in training, so you understand the technology
• Listen to native Castilian Spanish speakers in customer service contexts to internalize authentic pacing and tone choices

The facts

Pay

$50–100/hr

Hours

Hourly

Where

Remote

Open to

ESP, ESP

Field

Language and Audio

Posted

6/1/2026

Places left

7

We wrote this page from the public Mercor listing. It may be out of date, so read the full posting before you apply.