Website:
Job details:
Perit AI
Perit builds AI and ML training data for frontier labs. We work across audio, image, video and reinforcement learning environments, producing the high-quality, human-generated datasets the next generation of models is trained on.
We're hiring freelance annotators for transcription work. You'll listen to short clips of real conversational audio and write down exactly what's said - every word, filler, false start and repetition, following a detailed set of conventions.
This is verbatim transcription, not clean-up. The work is detail-heavy and quiet. If you like careful, focused tasks and have a good ear for language, it's a strong fit. Fully remote and fully flexible.
๐ช๐ต๐ฎ๐ ๐๐ผ๐'๐น๐น ๐ฏ๐ฒ ๐ฑ๐ผ๐ถ๐ป๐ด
- Listening to short audio clips and transcribing exactly what's spoken
- Capturing fillers, repetitions and cut-off words rather than tidying them away
- Marking speakers, overlapping speech, laughter, background noise and unclear audio using a defined tag set
- Writing numbers, acronyms and web addresses as they're spoken, not as they'd be written
- Applying detailed guidelines consistently across large volumes of work
- Flagging unclear audio, edge cases and anything the guidelines don't cover
๐ช๐ต๐ฎ๐ ๐๐ผ๐'๐น๐น ๐ป๐ฒ๐ฒ๐ฑ
- A laptop or desktop. This work can't be done on a phone or tablet.
- Headphones and a quiet environment โ you'll replay short segments repeatedly
- Strong English writing and comprehension, and comfort with detailed guidelines
- Patience for replaying the same two seconds until you're certain what you heard
No prior transcription experience required. Training and guidelines are provided.
๐๐ผ๐ ๐๐ผ ๐ฎ๐ฝ๐ฝ๐น๐
Fill out this google form (compulsory) If we see a fit, we'll reach out by email with an assessment. Pass it and you're in.
https://forms.gle/APvrbzMMK6UdMjkj7
Click on Apply to know more.