Research Transcriber

Princeton University
Princeton, NJ 08542
Sat Jul 02 2022 03:53:59 GMT+0000 (Coordinated Universal Time) save share
The Handwritten Text Recognition project at the Princeton Geniza Lab aims to train a computer model to recognize and automatically transcribe texts from the Cairo Geniza, a cache of Hebrew-script texts discovered in a medieval Egyptian synagogue. Researchers are responsible for improving the algorithm that recognizes Hebrew script by reviewing the computer’s auto-detection of the location of Hebrew script and ensuring that all areas on a page that contain Hebrew are correctly identified and delineated. This includes identifying the direction of text, marking each line of text from beginning to end, and correctly classifying any interlinear corrections. Researchers then examine the Hebrew-script text on the fragment, comparing it visually to an automatically uploaded transcription to ensure that the algorithm identifies each part of the image with its correct transcription. Researchers review the transcription line-by-line, ensuring that it accurately reflects the script written on the page. If there are discrepancies, researchers consult records of transcriptions as well as the auto-transcript to locate the correct section for each line of text.

This is a one-year term position and the end date is one-year from start date.
  • Segmenting text (reviewing the computer's automatic detection of Hebrew script to ensure that all Hebrew script on a page is correctly identified and its location pinpointed with specificity); Consulting database records as necessary to confirm the classification of text; Attending weekly team meetings and remaining current with policies and protocols.
  • Aligning transcriptions (visual comparison of the Geniza documents to the automatically uploaded transcriptions to ensure that the text accurately reflects the script written on the page); Consulting database records or historical experts as necessary to resolve discrepancies between the automatic and human-produced transcriptions.
  • Performing quality assurance: reviewing documents aligned by other researchers to ensure that all documents adhere to the same conventions consistently (for greater algorithmic accuracy).
Essential qualifications include:
  • Bachelor's degree with one year experience
  • Ability to read Hebrew script (Hebrew fluency preferred)
  • In-progress undergraduate studies or completed Bachelor’s in a related field (e.g., Near Eastern Studies, Judaic Studies, History or Computer Science)
  • Strong motivation to engage with Middle Eastern cultures and texts
  • Comfort and ease in learning new technological tools
  • Ability to follow a streamlined workflow accurately
  • Excellent attention to detail
  • Strong communication skills
  • Reliability and responsibility
Preferred knowledge includes:
  • Familiarity with Hebrew scribal scripts
  • Familiarity with the corpus of classical Jewish literature
Princeton University is an Equal Opportunity/Affirmative Action Employer and all qualified applicants will receive consideration for employment without regard to age, race, color, religion, sex, sexual orientation, gender identity or expression, national origin, disability status, protected veteran status, or any other characteristic protected by law. EEO IS THE LAW
Standard Weekly Hours: 20.00 Eligible for Overtime: Yes Benefits Eligible: Yes Probationary Period: 90 days Essential Services Personnel (see policy for detail): No Comments Related to End Date: One year term - end date one year from start date Physical Capacity Exam Required: No Valid Driver's License Required: No : #LI-NC1