Processing Language Resources of Under-Resourced and Endangered Languages for the Generation of Augmentative Alternative Communication Boards

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 2644–2648 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 2644 Processing Language Resources of Less Resourced and Endangered Languages for the Generation of Augmentative Alternative Communication Boards Anne Ferger University of Hamburg Germany anne.ferger@uni-hamburg.de Abstract Under-resourced, under-studied and endangered or small languages yield problems for automatic processing and exploiting because of the small amount of available data as well as the missing or sparse description of the languages. This paper shows an approach using different annotations of enriched linguistic research data to create communication boards commonly used in Alternative Augmentative Communication (AAC). Using manually created lexical analysis and rich annotation (instead of high data quantity) allows for an automated creation of AAC communication boards. The example presented in this paper uses data of the indigenous language Dolgan (an endangered Turkic language of Northern Siberia) created in the project INEL(Grammatical Descriptions, Corpora and Language Technology for Indigenous Northern Eurasian Languages) (Arkhipov and D¨abritz, 2018) to generate a basic communication board with audio snippets to be used in e.g. hospital communication or for multilingual settings. The created boards can be imported into various AAC software. In addition, the usage of standard formats makes this approach applicable to various different use cases. Keywords: language resources, endangered languages, augmentative alternative communication 1. Introduction (Re-)Using cost-intensively generated language resources is a question of sustainability as well as economic ef- ficiency. By exploiting the advantages of enriched lan- guage resources created in linguistic research projects while working with standard formats, a more wide-spread use of the language resources can be achieved. The approach described in this paper uses a standard format of linguistic transcription data of an endangered language and generates standard format communication boards with audio snippets to be imported into various alternative aug- mentative communication (AAC) software. AAC includes various forms of communication for people with differ- ent limitations to their communication abilities. This in- cludes low-tech (such as card-board communication boards or books with symbols for individuals to point at for com- munication) as well as high-tech (such as voice output com- munication aids (VOCAs) with eye-tracking functionality) solutions. The choice of AAC devices and their application is heavily dependent on the individual needs and abilities of the users. Simple communication is also often used in hos- pital settings where the ability to communicate can e.g. be temporarily obstructed by medical devices. Furthermore, simple communication boards can be used in multilingual settings where emergency basic communication is needed 1 . The INEL Dolgan Corpus (D¨abritz et al., 2019) used for this task is available under the licence CC BY-NC-SA 4.0 (public)2. Dolgan, a Turkic language spoken on the Taymyr peninsula and in adjacent areas in Northern Siberia has ap- proximately 1,000 speakers. 1e.g. in refugee camps https://www.tobiidynavox.com/support-training/downloads/ boardmaker/refugee-communication-boards/ 2PID:http://hdl.handle.net/11022/0000-0007-CAE7-1 While the example communication board created in this pa- per functions as a proof of concept and its real life usage may not be widespread, the …

Content truncated to 500 words