Voice & Voicemail Recording — Native Speakers (EU)

About OpenTrain

OpenTrain is the #1 platform for finding and building careers in AI training and data labeling. We connect people with hands-on projects that help teach and improve AI systems, and we make it easy to apply, build a profile, and start contributing.

About AI training work

AI training (also called data labeling or human feedback) is the human side of building artificial intelligence. Projects range from annotating images and transcribing audio to recording voice samples and rating model responses. This role helps improve IVR and call‑routing systems by providing real phone greetings and voicemail samples.

The role — quick project summary

You will record approximately 30 short audio clips (each up to 30 seconds): about 15 scripted phone greetings plus roughly 15 operator/voicemail clips. Exact counts depend on the operator message type assigned.

This is remote, flexible contract work: most participants complete the full batch in 30–45 minutes. The fixed payout for each batch is $24–$36, depending on assignment details.

• Data type: audio recordings for IVR/call‑routing model training.
• Employment: contractor, part‑time; entry‑level.
• Time: typically under 20 hours/week; one short batch per assignment.

Languages and eligible locations

We need native speakers located in the corresponding country for each language. Eligible language-country pairs are:

Flemish (Belgium), Danish (Denmark), French (France), German (Germany), Irish / English (Ireland), Italian (Italy), Dutch (Netherlands), Portuguese (Portugal), Spanish (Spain), English (United Kingdom).

What you'll do

Follow a short qualification process, then record the assigned batch using our web tool while connected to your in‑country mobile network. Recordings must be original, natural, and made in a quiet environment.

• Submit 2–3 short qualification clips to verify audio quality before the full batch.
• Record ~15 scripted greetings and ~15 operator/voicemail clips (each ≤30 seconds) in the web tool.
• Use your in‑country mobile number/SIM (no VoIP) to place calls as instructed.
• Sign the required consent to allow use of recordings for model training and QA.
• One account per person; do not use voice changers or text‑to‑speech.

Requirements — must-haves

All requirements below are mandatory for consideration. Candidates who do not meet these will be rejected during qualification.

• Native speaker of the chosen language and currently located in the corresponding country.
• 18 years or older and able to sign consent for use of your voice recordings.
• Have an in‑country mobile number or SIM (not VoIP) and be able to place calls on the local cellular network.
• Smartphone with a working microphone and access to a quiet recording environment.
• Can complete the full batch of ~30 recordings within 24–48 hours of assignment.
• Agree to provide original recordings and submit only one account.

How the qualification and recording process works

Apply and select the language you are applying for. We will confirm your country, native status, and device/network readiness through a short qualification checklist. If you pass, you'll submit 2–3 short qualification clips and then receive the full assignment.

All recordings are captured in our web tool while using your in‑country mobile network. After you submit the batch, recordings will be reviewed for quality and compliance with the consent terms.

• You may be asked for your mobile carrier (e.g., Orange, Vodafone, Deutsche Telekom) to help route calls.
• Optional: you can share gender and regional/dialect info to help dataset balance (not required).

Who should apply

This project is ideal for native speakers who want short, flexible remote work and are comfortable recording brief phone-style phrases. No prior annotation experience is required — attention to clarity, following prompts, and meeting device/network requirements are the key skills.