Literary Transcription and Verbatim Dictation Accuracy
Literary Transcription and Verbatim Dictation Accuracy
This domain tests the reporter's ability to capture speech exactly under dictation and in literary or narrative passages, then render it with faithful punctuation, spelling, and speaker identification. Accuracy includes what is said, how it is said, and what cannot be heard.
Core rules
- Verbatim means every word, repetition, false start, and completed correction is preserved unless the court orders otherwise. Do not polish grammar or delete profanity.
- Mark non-verbal events with parentheticals: (Laughter), (Pause), (Witness crying), (Exhibit marked). Use them consistently and only for events that affect the record.
- Identify speakers at every change; in depositions, use Q and A; in hearings, use titles such as THE COURT, MR. SMITH, MS. JONES.
- For dictation, write in phrases, maintain rhythm, and read back from the machine or CAT screen. If a word is unclear, leave a placeholder and verify from audio rather than inventing.
- Use [inaudible 00:14:22] or a comparable time-stamped note when speech cannot be recovered. Do not substitute a likely word.
- Preserve dialect and nonstandard forms in quotation; do not translate them into standard English.
Worked example
Audio: I-I didn't-no, I didn't see him. A weak transcription reads: I didn't see him. The accurate version is: I-I didn't-no, I didn't see him. Reasoning: the interrupted start, self-correction, and repetition are part of the testimony and may affect credibility. The dashes mark abrupt breaks; the words remain unchanged. If a cough obscures the middle, the reporter inserts (Cough) and flags any unrecoverable word rather than guessing.
Traps
- Cleaning up false starts to make a smoother sentence.
- Guessing a name or technical term from context.
- Using ellipses to hide uncertain speech.
- Omitting non-verbal events that the judge or parties may later dispute.