Full-Time Verbatim Transcriptionist

We are seeking highly accurate, dependable verbatim transcriptionists to proofread, correct, and format AI-generated transcripts. The transcriptionist will listen carefully to the original audio or video recording, compare it against the AI-generated transcript, correct all errors, and deliver a professionally formatted transcript with an accuracy level of 99% or higher. This position requires excellent listening skills, strong English grammar and punctuation knowledge, careful attention to detail, and the discipline to never guess at unclear speech. When speech cannot be understood after careful review, the transcriptionist must insert the required inaudible tag with an accurate time code instead of guessing, assuming, or entering a potentially misheard word. Primary Responsibilities - Listen to audio or video recordings while reviewing an AI-generated transcript. - Correct misheard words, omitted content, added content, speaker-label errors, punctuation errors, and formatting problems. - Produce an accurate verbatim transcript that reflects what each speaker actually said. - Preserve repetitions, false starts, filler words, grammatical errors, and other spoken-language features when full-verbatim transcription is required. - Identify speakers accurately and apply consistent speaker labels. - Insert properly formatted inaudible tags with time codes, such as [inaudible 00:03:27], whenever speech cannot be understood. - Never guess, reconstruct, or insert words that cannot be verified from the recording. - Research names, organizations, technical terms, locations, and other verifiable information when necessary. - Follow all project-specific transcription, formatting, and style requirements. - Format headings, speaker labels, paragraphs, time codes, annotations, and other transcript elements consistently. - Proofread the entire completed transcript against the source recording before submission. - Deliver transcripts that meet or exceed a 99% accuracy standard. - Meet assigned deadlines and communicate promptly about audio-quality or source-file problems. - Maintain strict confidentiality when handling private or sensitive recordings. Accuracy Requirements The AI-generated transcript is only a starting point. Every word must be reviewed carefully against the original recording. The final transcript must not contain: - Guessed words presented as confirmed speech - Misheard words that were not corrected - Unmarked unintelligible or inaudible speech - AI-generated wording that was not spoken - Missing statements or omitted speaker turns - Incorrect speaker identifications - Improper or missing time codes - Inconsistent formatting - Avoidable spelling, grammar, or punctuation errors When a word or passage cannot be understood confidently, the transcriptionist must use the required inaudible tag with a time code. Guessing is not acceptable. Required Qualifications - Demonstrated experience with verbatim transcription, transcript editing, or professional proofreading - Ability to achieve and consistently maintain at least 99% accuracy - Excellent command of English spelling, grammar, punctuation, and sentence structure - Strong listening comprehension - Ability to distinguish between multiple speakers - Ability to identify and correct misheard words in AI-generated transcripts - Ability to recognize when speech is unclear and use an inaudible tag instead of guessing - Experience proofreading and correcting automated or AI-generated transcripts - Ability to follow detailed style guides and project instructions - Strong research skills for confirming names, terminology, and contextual information - Ability to work independently and meet firm deadlines - Reliable computer, headphones, word-processing software, and high-speed internet connection - Commitment to confidentiality and secure file handling Preferred Qualifications - Experience transcribing legal, government, corporate, academic, medical, insurance, investigative, or public-meeting recordings - Experience working with recordings containing multiple speakers, overlapping speech, background noise, accents, or specialized terminology - Familiarity with Microsoft Word and professional transcript-formatting requirements - Experience using transcription software, audio playback controls, foot pedals, or time-code tools - Familiarity with clean-verbatim and full-verbatim transcription conventions Hiring Assessment As part of the hiring process, qualified applicants will be required to complete a transcription assessment using an audio or video file that is at least three minutes long. The assessment may require the applicant to: - Review and correct an AI-generated transcript. - Compare the entire transcript against the original recording. - Correct all misheard words, omissions, additions, speaker-label errors, punctuation errors, and formatting problems. - Preserve the required level of verbatim detail. - Insert inaudible tags with accurate time codes where speech cannot be understood. - Format the transcript according to the provided instructions. - Proofread the completed transcript against the recording before submission. The assessment will be evaluated based on: Overall transcription accuracy - Ability to identify and correct AI-generated errors - Ability to avoid guessing - Correct use of inaudible tags and time codes - Verbatim completeness - Speaker-label accuracy - Grammar, punctuation, and spelling - Formatting consistency - Attention to instructions - Final proofreading quality Completion of the assessment does not guarantee an offer of work. All assessment recordings, transcripts, and related materials are confidential and may not be copied, shared, published, or distributed. Please submit: - Your résumé or a summary of your transcription experience - The types of recordings you have transcribed Your estimated transcription or transcript-editing speed - Your general availability - A brief explanation of what you do when a word or passage cannot be understood - Any relevant accuracy scores, certifications, or work samples Applicants selected for the next stage will receive instructions for the required three-minute-or-longer transcription assessment. Important: Applicants who guess at unclear words instead of using properly time-coded inaudible tags will not meet the position’s accuracy requirements.