OAK – Display Transcription and Disfluency Tagging

About this role

Turn everyday speech into clean, accurate written text in your language. Flexible, remote, task-based work for people with a sharp ear and an eye for detail. OAK is a multilingual transcription project made up of two simple, connected steps. First you transcribe short clips of natural speech into accurate written text. Then you review your own transcripts and tag the small slips that show up in real conversation, such as repeated words, filler sounds, and mid-sentence corrections. The result is clean, reliable text that helps improve the speech and language tools people use every day. Work is released by the project team in manageable batches, so you complete tasks when it suits your schedule. There is no commute, no fixed shift, and no prior experience required beyond strong language skills and attention to detail. What You’ll Do Display Transcription. Listen to short speech recordings in your language and type exactly what is said, following clear project guidelines. You capture the words as spoken, without adding, removing, or reordering anything. Apply the writing conventions. Turn spoken numbers, dates, times, currency, percentages, measurements, phone numbers, web addresses, and email addresses into their standard written form. Keep spelling, capitalization, punctuation, names, and accents correct for your locale. Use the required tags. Mark the elements the guidelines ask for, accurately and consistently. Disfluency Tagging. Once a transcript is finalized, identify and tag four common types of speech slips: meaningless repetitions, empty filler words and sounds, unnecessary informal markers, and mid-speech self-corrections. You learn to tell a genuine connecting word apart from filler, so nothing gets tagged by mistake. Polish the result. Tidy up punctuation and capitalization so the text reads cleanly once the slips are removed, following the project rules for tagging, merging, and boundaries. Who We’re Looking For Native or near-native fluency i