German Audio Recording Expert

$20–40/hrrate as listed

Record scripted and spontaneous German speech for AI training and evaluation datasets, following per-batch specs for microphone setup, room noise, and delivery style. Work is remote and asynchronous, paid per accepted recording or hourly, with observed rates in the $20–40/hr range depending on task complexity and dialect need.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.

What the work involves

You record German audio against detailed collection specs. A batch might ask for 200 read prompts at conversational pace, spontaneous monologues on assigned topics, call-center-style role plays, or wake-word and command utterances repeated at varying distances from the mic. Specs usually fix the sample rate, file format, naming convention, and acceptable noise floor, and sometimes require deliberate variation — faster delivery, background TV, phone-held-at-arm's-length.

Beyond recording, contributors are often asked to self-QA: listen back, flag clipping, breath pops, misread words, or timing drift against the transcript; and occasionally review other contributors' submissions against the same rubric. Rejections are normal early on and usually trace back to room reverb, inconsistent gain, or drifting from the requested register rather than to language ability.

What the platform screens for

Native or near-native German, with a clear picture of your regional accent and whether you can hold a neutral Standarddeutsch register on request.
Recording hygiene — that you own or can borrow a usable mic, know what a quiet room actually sounds like, and can produce WAV files at a specified rate without resampling artifacts.
Instruction fidelity, tested through follow-ups: what you would do when a prompt contains a typo, an anglicism, or a sentence no native speaker would say aloud.
Consistency across sessions, since datasets break when your distance, gain, or energy changes between day one and day five.

Logistics

Fully remote and asynchronous. Batches are released with deadlines rather than shifts, so you choose your recording hours — though a genuinely quiet time of day matters more here than in text work. Most contributors take 5–20 hours a week; volume is irregular and tied to client collection cycles. Expect an onboarding batch that is graded before larger volume is released.

What they look for

Audio recording

Before you apply

Would you pass this screen?

Drawn from what this listing actually requires. Answered in your browser — nothing is sent to us, and we do not store your answers.

Are you a native speaker of German, or have you been fluent in German for at least 5 years?
Native or near-native German · required
Do you own a USB microphone, XLR mic with interface, or documented access to equivalent equipment suitable for voice recording?
Microphone and recording setup · required
Do you have consistent access to a quiet indoor space (bedroom, closet, or similar) where you can record without significant background noise, traffic, or HVAC hum?
Quiet recording environment · required
Can you record and export WAV files at specified sample rates (e.g., 48 kHz) using free or paid audio software?
DAW or recording software · required
Can you commit to working asynchronously on batches released with deadlines, rather than on fixed shifts, with the understanding that volume varies by client cycles?
Availability for deadline-based work · preferred

0 of 5 answered

Included with this listing

Interview prep — 10 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: Whether you own or can reliably access a working microphone, understand acoustic fundamentals well enough to identify and work around room problems, and can actually produce files at a specified sample rate without upsampling artifacts.
A strong answer: Name a specific room (bedroom, closet, kitchen corner) and explain how you'd test it for noise—ambient hum, traffic, HVAC—using free software or hardware meters; name the microphone model or class you'd use; and describe the export workflow in your DAW or recording app to guarantee 48 kHz without resampling.
Where people lose it: Claiming a 'quiet room' without any concrete method to verify it, describing a smartphone or earbuds mic as adequate, or remaining vague about file export—'I'll just save it as WAV' fails if your audio editor resamples on export and you don't catch it.