What the work involves
This is a single-session audio contribution task, not an ongoing role. AfterQuery supplies a short script and a written recording guide; you read the script aloud in your own voice, record it on a phone or computer microphone, and upload the file through the platform. Start to finish, contributors report about 10–15 minutes. The flat rate is $50 per accepted submission, which is where the observed $30/hr equivalent comes from — the hourly figure is a comparison, not a promise of ongoing hours.
The purpose is accent coverage. Speech models are routinely weaker on Bengali-influenced English than on the accents that dominate training corpora, and the fix is authentic recorded speech from people who actually speak that way. That is why the listing is explicit about no imitated accents — a performed accent contaminates the dataset it is meant to correct.
What the platform screens for
- Authenticity of accent. You speak English with a natural Bengali influence because that is how you speak, not as a performance. Expect a question about your language background and where you learned English.
- Recording conditions. A quiet room, a working microphone, no background television or traffic. Nothing studio-grade is required; a modern smartphone in a closed room is fine.
- Instruction-following. The guide specifies file format, distance from the mic, and how to handle mistakes. Submissions that ignore those specs get rejected, and rejection means no payment.
- Clean reading. Reading the script as written, at a natural pace, without ad-libbing or skipping lines.
Logistics
Fully remote and fully asynchronous. There is no interview call, no scheduled shift, and no minimum ongoing commitment. Quick turnaround is the point — contributors are generally asked to submit within a day or two of receiving the script. Payment is per accepted submission through the platform's standard payout process. Occasional repeat invitations happen when new scripts open, but nothing about this listing guarantees repeat work.