CHA file parser for Python

Class for parsing CLAN's CHA file.

Made by Leandro Garber from CIIPME-CONICET

Features

• Utterances as a list of strings
• MOR tier as objects
• Easily add more custom tiers
• Count tokens and types of words, utterances, nouns, verbs and adjectives. Filter by child directed, child produced and overheard speech.
• Count main verbs, either referring to physical or mental actions. Auxiliary verbs present in periphrastic verbs are excluded. (spanish only)

Usage

Import

import sys
sys.path.insert(0, '<path_to_cloned_repo>')

from ChaFile import *

Instance

cha = ChaFile(<path_to_cha_file>)

Options

cha = ChaFile(<path_to_cha_file> )

Get utterances

lines = cha.getLines()

Each line is an object with:

• LINE_UTTERANCE : The text of the utterance
• LINE_NUMBER
• LINE_SPEAKER
• LINE_ADDRESSEE
• LINE_BULLET : Timestamp
• TIER_MOR : A list of objects with MOR data: MOR_UNIT_LEXEMA and MOR_UNIT_CATEGORIA
• ... any other tier

Cite

Garber, L. (2019). CHA file python parser. Zenodo. https://doi.org/10.5281/zenodo.3364020