NYPL jobs come from the New York Public Library. These jobs generally feature two speakers conversing over a virtual call. Usually, there is an interviewer and an interviewee, discussing various topics, often the history of dance. The jobs are focused on the interviewee's stories and first-hand experience for the given topic. 

NYPL jobs have a unique set of instructions compared to other Custom jobs. These jobs DO require sound markers for any nonverbal sounds such as laughing, crying, sighing, et cetera. These instances should be captured with a bracketed lowercase sound marker on its own line. NYPL has specifically requested that we capture as much of the essence of the conversation as possible, and we're required to add a marker for any situation where the audio is playing into the context of the given dialogue. 

JOB INSTRUCTIONS:

• Line formatting: Paragraph. Create a new paragraph for each new speaker. We can also use discretionary line breaks with a single speaker for long pauses or topic changes.  
   
• Verbatim: Transcribe verbatim at all times including: false starts, stutters ("I — I — I"), nonfluencies (um, uh, erm), repetitions, habitual speech and fillers (like, I mean, you know). As usual, do include interjections and any expressions. In short, strive to capture ALL the audio!  
   
• Speaker Labels: Indicate the names of each speaker with each speaker change in all CAPS. Use FULL NAME for the first instance, LAST NAME only in all following instances.
  
  JOHN MALOY: Welcome to the first TED Talk.
  DIANA SMITH: Thanks for having me.
  MALOY: Hope you’re having a great day.
  SMITH: Yes, I certainly am.
  
  There may be times when you dock the middle portion of a job. In these cases, since the full-name IDs have likely been created in an earlier segment, you can treat the IDs as last name only. Please contact a CSL if you're not sure. 
   
• Interruptions: Use a double dash (--) for interruptions or stutters, et cetera. Do NOT use an em dash or an ellipsis.
   
• Sound Markers and Audio Description: Add lowercase, bracketed audio description wherever necessary. Please use these markers on their own line. These markers will vary depending on what's needed to help visually understand the context. 
  
  Here's some examples of sound markers and how they interact with the job text. Of course, you can create your own if the context of the job requires it. 
  
   
  
  Generally, no new ID is needed after a sound marker, unless the sound marker creates an elongated pause and/or the sound marker includes some type of overlapping audio where the last person to speak was unclear. 
  
   

Consider the following examples:

 

The two different markers create a totally different emotional response to the phrase. The first would be understood as playful, joking and silly. But in the second example, the exact same dialogue now becomes more somber. It's our job on Custom to capture these instances in the transcript as warmly and accurately as possible.