What the work involves
This is a single-task audio contribution, not an ongoing role. You receive a short English script and a written recording guide, record yourself reading it on a phone or computer microphone, and upload the file through the designated submission platform. The whole cycle is designed to take about fifteen minutes, and the posting states a flat $50 for a single accepted submission — the $30/hr band shown on this board is a normalised estimate, not a rate you are paid by the hour. Repeat batches sometimes follow when a dataset needs expansion, but nothing here is promised as recurring.
What the platform screens for
The core screen is authenticity of accent. AfterQuery is building speech data that represents how Gujarati-first English speakers actually sound, so an imitated, performed, or coached-neutral accent defeats the purpose and is grounds for rejection. Expect to be asked directly about your language background — where you grew up, which language you speak at home, how long you have spoken English — and expect a short spoken sample. The second screen is procedural: can you read a written guide once and produce a file that meets the spec without back-and-forth. Common failures are background noise, clipping from holding the mic too close, wrong file format, and reading the script inaccurately or adding filler.
Practical logistics
- Fully remote and fully asynchronous; no calls, no scheduled shifts.
- You need a quiet room and a working microphone — phone handset mics are acceptable if the guide says so.
- Turnaround is short, typically same-day or next-day once the script lands.
- Payment is per accepted submission; recordings that fail QA for noise or non-compliance may be sent back for one retake.
This suits someone who wants a small, clearly bounded paid task and may want to be on the list when similar language-data projects open, rather than someone looking for steady weekly hours.