ddhi-encoder 1.2.0

Encoding tools for DDHI

A collection of command-line utilities to assist in the creation of TEI-encoded oral history interviews. Part of the Dartmouth Digital History Initiative.

DDHI Encoder

The ddhi-encoder package is being developed to assist encoders in the DDHI project in encoding oral history interview transcripts in TEI. At present, it contains three command-line utilities:

1. ddhi_convert: convert a Dartmouth DVP transcript from docx to tei.xml.

2. ddhi_tag: perform named-entity tagging on a DDHI TEI transcription.

3. ddhi_mentioned_places: extract places from stand-off markup for processing with OpenRefine

4. ddhi_update_places: update places in stand-off markup

Installation

You can use pip to install this package:

pip install ddhi-encoder

To peform named-entity tagging with ddhi_tag, you will need a Spacy model. Before running ddhi_tag, install Spacy’s small English model:

python -m spacy download en_core_web_sm

See the Spacy documentation for more information.

Use

Use ddhi_convert to transform a DOCX-encoded transcription into a simply structured TEI document:

ddhi_convert ~/Desktop/transcripts/zien_jimmy_transcript_final.docx -o tmp.tei.xml

Use ddhi_tag to add named-entity tags to a TEI-encoded transcription:

ddhi_tag -o zien.tei.xml tmp.tei.xml

Use ddhi_mentioned_places to extract the places in a TEI file’s standoff markup and print it as tab-separated values:

ddhi_mentioned_places lovely.tei.xml > lovely.tsv

Use ddhi_update_places to update the places in a TEI file’s standoff markup with identifiers and geo-coordinates obtained via OpenRefine or other procedure:

ddhi_update_places lovely.tei.xml lovely_updates.tsv >
updated_lovely.tei.xml