The ǂKhomani San, Hugh Brody Collection features the voices and history of indigenous hunter gatherer descendants in three endangered languages namely, N|uu, Kora and Khoekhoe as well as a regional dialect of Afrikaans. A large component of this collection is audio-visual (legacy media) recordings of interviews conducted with members of the community by Hugh Brody and his colleagues between 1997 and 2012, referring as far back as the 1800s. The Digital Library Services team at the University of Cape Town aim to showcase the collection digitally on the UCT-wide Digital Collections platform, Ibali which runs on Omeka-S. In this paper we highlight the importance of such a collection in the context of South Africa, and the ethical steps that were taken to ensure the respect of the ǂKhomani San as their stories get uploaded onto a repository and become accessible to all. We will also feature some of the completed collection on Ibali and guide the reader through the organisation of the collection on the Omeka-S backend. Finally, we will outline our development process, from digitisation to repository publishing as well as present some of the challenges in data clean-up, the curation of legacy media, multi-lingual support, and site organisation.

This contribution describes a free and open mobile dictionary app based on open dictionary data. A specific focus is on usability and user-adequate presentation of data. This includes, in addition to the alphabetical lemma ordering, other vocabulary selection, grouping, and access criteria. Beyond search functionality for stems or roots – required due to the morphological complexity of Bantu languages – grouping of lemmas by subject area of varying difficulty allows customization. A dictionary profile defines available presentation options of the dictionary data in the app and can be specified according to the needs of the respective user group. Word embeddings and similar approaches are used to link to semantically similar or related words. The underlying data structure is open for monolingual, bilingual or multilingual dictionaries and also supports the connection to complex external resources like Wordnets. The application in its current state focuses on Xhosa and Zulu dictionary data but more resources will be integrated soon.

The recent advances in Natural Language Processing have only been a boon for well represented languages, negating research in lesser known global languages. This is in part due to the availability of curated data and research resources. One of the current challenges concerning low-resourced languages are clear guidelines on the collection, curation and preparation of datasets for different use-cases. In this work, we take on the task of creating two datasets that are focused on news headlines (i.e short text) for Setswana and Sepedi and the creation of a news topic classification task from these datasets. In this study, we document our work, propose baselines for classification, and investigate an approach on data augmentation better suited to low-resourced languages in order to improve the performance of the classifiers.

Setswana language is one of the Bantu languages written disjunctively. Some of its parts of speech such as qualificatives and some adverbs are made up of multiple words. That is, the part of speech is made up of a group of words. The disjunctive style of writing poses a challenge when a sentence is tokenized or when tagging. A few studies have been done on identification of multi-word parts of speech. In this study we go further to tokenize complex parts of speech which are formed by extending basic forms of multi-word parts of speech. The parts of speech are extended by recursively concatenating more parts of speech to a basic form of parts of speech. We developed rules for building complex relative parts of speech. A morphological analyzer and Python NLTK are used to tag individual words and basic forms of multi-word parts of speech. Developed rules are then used to identify complex parts of speech. Results from a 300 sentence text files give a performance of 74%. The tagger fails when it encounters expansion rules not implemented and when tagging by the morphological analyzer is incorrect.

In this paper, we compare four state-of-the-art neural network dependency parsers for the Semitic language Amharic. As Amharic is a morphologically-rich and less-resourced language, the out-of-vocabulary (OOV) problem will be higher when we develop data-driven models. This fact limits researchers to develop neural network parsers because the neural network requires large quantities of data to train a model. We empirically evaluate neural network parsers when a small Amharic treebank is used for training. Based on our experiment, we obtain an 83.79 LAS score using the UDPipe system. Better accuracy is achieved when the neural parsing system uses external resources like word embedding. Using such resources, the LAS score for UDPipe improves to 85.26. Our experiment shows that the neural networks can learn dependency relations better from limited data while segmentation and POS tagging require much data.

Linguistic fieldworkers collect and archive metadata as part of the language resources (LRs) that they create, but they often work in resource-constrained environments that prevent them from using computers for data entry. In such situations, linguists must complete time-consuming and error-prone digitization tasks that limit the quantity and quality of the resources and metadata that they produce (Thieberger & Berez 2012; Margetts & Margetts 2012). This paper describes a method for entering linguistic metadata into mobile devices using the Open Data Kit (ODK) platform, a suite of open source tools designed for mobile data collection.

The paper describes aspects of an HPSG style computational grammar of the West African language Ga (a Kwa language spoken in the Accra area of Ghana). As a Volta Basin Kwa language, Ga features many types of multiverb expressions and other particular constructional patterns in the verbal and nominal domain. The paper highlights theoretical and formal features of the grammar motivated by these phenomena, some of them possibly innovative to the formal framework. As a so-called deep grammar of the language, it hosts a rich lexical structure, and we describe ways in which the grammar builds on previously available lexical resources. We outline an environment of current resources in which the grammar is part, and lines of research and development in which it and its environment can be used.

Creating a new wordnet is by no means a trivial task and when the target language is under-resourced as is the case for the languages currently included in the multilingual African Wordnet (AfWN), developers need to rely heavily on human expertise. During the different phases of development of the AfWN, we incorporated various methods of fast-tracking to ease the tedious and time-consuming work. Some methods have proven effective while others seem to have little positive impact on the work rate. As in the case of many other under-resourced languages, the expand model was implemented throughout, thus depending on English source data such as the English Princeton Wordnet (PWN) which is then translated into the target language with the assumption that the new language shares an underlying structure with the PWN. The paper discusses some problems encountered along the way and points out various possibilities of (semi) automated quality assurance measures and further refinement of the AfWN to ensure accelerated growth. In this paper we aim to highlight some of the lessons learnt from hands-on experience in order to facilitate similar projects, in particular for languages from other African countries.

In a context where open-source NLP resources and tools in African languages are scarce and dispersed, it is difficult for researchers to truly fit African languages into current algorithms of artificial intelligence. Created in 2017, with the aim of building communities of voluntary contributors around African native and/or national languages, cultures, NLP technologies and artificial intelligence, the NTeALan association has set up a series of web collaborative platforms intended to allow the aforementioned communities to create and administer their own lexicographic resources. In this article, we present on the one hand the first versions of the three platforms: the REST API for saving lexicographical resources, the dictionary management platform and the collaborative dictionary platform; on the other hand, we describe the data format chosen and used to encapsulate our resources. After experimenting with a few dictionaries and some users feedback, we are convinced that only collaboration-based approach and platforms can effectively respond to the production of good resources in African native and/or national languages.