arXiv subjects

Howard Lakougna

Publications and source records attributed to Howard Lakougna.

3 recordsLinked to original sources

arXiv2026

WAXAL: A Large-Scale Multilingual African Language Speech Corpus

The advancement of speech technology has predominantly favored high-resource languages, creating a significant digital divide for speakers of most Sub-Saharan African languages. To address this gap, we introduce WAXAL, a large-scale, openly accessible speech dataset for 24 languages representing over 100 million speakers. The collection consists of two main components: an Automated Speech Recognition (ASR) dataset containing approximately 1,250 hours of transcribed, natural speech from a diverse range of speakers, and a Text-to-Speech (TTS) dataset with around 235 hours of high-quality, single-speaker recordings reading phonetically balanced scripts. This paper details our methodology for data collection, annotation, and quality control, which involved partnerships with four African academic and community organizations. We provide a detailed statistical overview of the dataset and discuss its potential limitations and ethical considerations. The WAXAL datasets are released at https://huggingface.co/datasets/google/WaxalNLP under the permissive CC-BY-4.0 license to catalyze research, enable the development of inclusive technologies, and serve as a vital resource for the digital preservation of these languages.

eess.AS↗

arXiv2025

Building AI-based advisory services for smallholder farmers: Technical learnings from the AIEP Initiative

We report technical learnings from five AI-based agricultural advisory MVPs deployed in Kenya and Bihar, India, under the AIEP Initiative. A 800-farmer study found high user satisfaction (NPS ~60). All solutions implement a modular two-part architecture: (i) an interface component (IVR /WhatsApp / app) with ASR-MT-TTS for multilingual voice access; and (ii) a reasoning component combining LLMs capabilities with query orchestration, external data (weather/soil/markets), and RAG over curated agricultural corpora. We describe key challenges: (a) latency, especially for voice; reductions were achieved via in-country hosting and audio minimization, but consistent <5s remains challenging; (b) language coverage: low-resource ASR/MT integration and nonstandard scripts hinder end-to-end quality; and (c) corpus curation: access, validation, and maintenance are labor-intensive, as well as provide recommendations on how to develop similar systems. We discuss common enablers including (a) data sharing, (b) common corpora, (c) better language AI and (d) evaluation and benchmarking. We also present golden Q&A sets to evaluate LLM capabilities for smallholder agriculture.

cs.HC↗

arXiv2025

Preliminary Ranking of WMT25 General Machine Translation Systems

We present the preliminary rankings of machine translation (MT) systems submitted to the WMT25 General Machine Translation Shared Task, as determined by automatic evaluation metrics. Because these rankings are derived from automatic evaluation, they may exhibit a bias toward systems that employ re-ranking techniques, such as Quality Estimation or Minimum Bayes Risk decoding. The official WMT25 ranking will be based on human evaluation, which is more reliable and will supersede these results. The official WMT25 ranking will be based on human evaluation, which is more reliable and will supersede these results. The purpose of releasing these findings now is to assist task participants with their system description papers; not to provide final findings.

cs.CL↗