arXiv subjects

Andrew Kipkebut

Publications and source records attributed to Andrew Kipkebut.

3 recordsLinked to original sources

arXiv2026

AfriVoices-KE: A Multilingual Speech Dataset for Kenyan Languages

AfriVoices-KE is a large-scale multilingual speech dataset comprising approximately 3,000 hours of audio across five Kenyan languages: Dholuo, Kikuyu, Kalenjin, Maasai, and Somali. The dataset includes 750 hours of scripted speech and 2,250 hours of spontaneous speech, collected from 4,777 native speakers across diverse regions and demographics. This work addresses the critical underrepresentation of African languages in speech technology by providing a high-quality, linguistically diverse resource. Data collection followed a dual methodology: scripted recordings drew from compiled text corpora, translations, and domain-specific generated sentences spanning eleven domains relevant to the Kenyan context, while unscripted speech was elicited through textual and image prompts to capture natural linguistic variation and dialectal nuances. A customized mobile application enabled contributors to record using smartphones. Quality assurance operated at multiple layers, encompassing automated signal-to-noise ratio validation prior to recording and human review for content accuracy. Though the project encountered challenges common to low-resource settings, including unreliable infrastructure, device compatibility issues, and community trust barriers, these were mitigated through local mobilizers, stakeholder partnerships, and adaptive training protocols. AfriVoices-KE provides a foundational resource for developing inclusive automatic speech recognition and text-to-speech systems, while advancing the digital preservation of Kenya's linguistic heritage.

cs.CL↗

arXiv2025

Building low-resource African language corpora: A case study of Kidawida, Kalenjin and Dholuo

Natural Language Processing is a crucial frontier in artificial intelligence, with broad applications in many areas, including public health, agriculture, education, and commerce. However, due to the lack of substantial linguistic resources, many African languages remain underrepresented in this digital transformation. This paper presents a case study on the development of linguistic corpora for three under-resourced Kenyan languages, Kidaw'ida, Kalenjin, and Dholuo, with the aim of advancing natural language processing and linguistic research in African communities. Our project, which lasted one year, employed a selective crowd-sourcing methodology to collect text and speech data from native speakers of these languages. Data collection involved (1) recording conversations and translation of the resulting text into Kiswahili, thereby creating parallel corpora, and (2) reading and recording written texts to generate speech corpora. We made these resources freely accessible via open-research platforms, namely Zenodo for the parallel text corpora and Mozilla Common Voice for the speech datasets, thus facilitating ongoing contributions and access for developers to train models and develop Natural Language Processing applications. The project demonstrates how grassroots efforts in corpus building can support the inclusion of African languages in artificial intelligence innovations. In addition to filling resource gaps, these corpora are vital in promoting linguistic diversity and empowering local communities by enabling Natural Language Processing applications tailored to their needs. As African countries like Kenya increasingly embrace digital transformation, developing indigenous language resources becomes essential for inclusive growth. We encourage continued collaboration from native speakers and developers to expand and utilize these corpora.

cs.CL↗

arXiv2022

Prototyping a Serial Number Based Authentication Model for a Computer in a Wireless Local Area Network

With the increase of wireless LAN usage in homes and enterprises due to its numerous benefits, authenticating the ever increasing number of devices and their users has become a challenge to proprietors of such kind networks. A MAC address, a physical network address that is used as basis for this study, has a copy of its value in the system software that can be spoofed and altered rendering the address not unique, not secure and unreliable. On the contrary, a computers serial number is hard-coded in the system hardware only and therefore cannot be spoofed and altered making it unique, secure and reliable. The research, therefore, was aimed at designing a model that demonstrates how a computers serial number can be used for authenticating a computer in a wireless local area network. In order to achieve the research objective, the study examined the inbuilt access and use of a computers serial number prototype model as an alternative method of authenticating devices in a network. Design science research methodology that involved design and development, demonstration and model evaluation was employed. A Serial Number Based Authentication Prototype or SNAP was therefore designed using state chart and flow chart diagrams based on dynamic programming, developed over evolutionary prototyping and test run on a static experimental design using Java Development Kit and MySQL platforms to demonstrate, as proof of concept, that a computers serial number can be used to authenticate a computer in a wireless local area network. From the test runs whose outcomes were the binary values yes or no, it was found out that SNAP can actually allow or deny, enable or disable a computer in a network based on the computers serial number. The researcher therefore, recommends that the prototype be scaled up, then adopted as a network device authentication method.

cs.CR↗