A Universal Dependencies corpus for Ika, a member of the Chibchan language family. The language is spoken by about 25,000 speakers in Colombia.
The treebank is an automatic conversion of complete trees of the SUD_Ika-ChibErgIS, which is an automatic conversion of the mSUD_Ika-ChibErgIS which was extracted from an interlinearized corpus in Flex format.
The original corpus consists of 40 texts recorded in the Ika language (41 recordings). The texts were collected as part of original fieldwork conducted between 2018 and 2022 in Pueblo Bello, Cesar, Colombia.
Sentences are annotated with the following metadata:
speaker_id(which identifies the turn of speech)sent_timecode(which will enable playback of the sentence)morphemic_text: (original segmentation of the text into morphemes)text: (lexical tokenization)text_en: (English interpretation)text_phrase-gls-de: (original id)text_phrase-gls-es: (Spanish interpretation)text_phrase-gls-tl: (original comments in Flex)
This version of the treebank is a dependency parsing of the original corpus first complete trees.
The original data are spoken data, which were originally segmented in words and interlinearized and glossed in Flex with concatenated clitics. Tokens comprize words and affixes (preceded by a "=" sign).
The corpus structure was designed by Stavros Skopeteas and comprises four different text types: spoken narratives (ancestor stories; AN), retellings of the Pear Story (PE), procedural texts (instructions for preparing a meal; CO), and comparative descriptions (contrasting the Ika people with other indigenous communities; CD). The labelling scheme, also developed by Stavros Skopeteas, follows a structured format to ensure that each text is uniquely identifiable and provides clear metadata for organisation. The format is as follows:
language-collection-instruction-00000-speaker
This label consists of the following components:
ARH: the language identifier, representing Ika in this corpus. TXT: indicates the "Texts" subcollection, showing that the recording belongs to the textual part of the corpus. AN, TS, CD, or PE: specify the type of recording. 00000: a placeholder component that has no functional role in this subcollection. The final number identifies the speaker, corresponding to a specific individual.
The UD_IKA-ChibErgIS counts 3,703 tokens for 428 sentences.
(none available yet)
This treebank was created as part of the ChibErgIS project.
Special thanks go to Bruno Guillaume for the SUD-to-UD conversion, and to Sylvain Kahane and Aleksandra Miletic for their support. I am also deeply grateful to Natalia Cáceres Arandia for her assistance with the (S)UD annotation, and to Florian Deichsler for his help with glossing in FLEx and solving a wide range of technical issues.
Finally, I would like to express my heartfelt thanks to the Ika native speakers for allowing me to record them, and especially to Leidy Karina Izquierdo Mejía and Gunyan Giovanny Hernán Izquierdo Mejía for transcribing the Ika data and assisting with the Spanish translations.
- 2025-05-15 v2.16
- Initial release in Universal Dependencies.
=== Machine-readable metadata (DO NOT REMOVE!) ================================ Data available since: UD v2.16 License: CC BY-SA 4.0 Includes text: yes Parallel: no Genre: spoken Lemmas: manual native UPOS: manual native XPOS: not available Features: manual native Relations: manual native Contributors: Bajorat, Jana; Cáceres Arandia, Natalia Contributing: here Contact: jana.bajorat@hu-berlin.de ===============================================================================