AfroScope: A Framework for Studying the Linguistic Landscape of Africa

TL;DR

AfroScope tackles the challenge of language identification in Africa's highly diverse linguistic landscape by building AfroScope-Data with 713 languages and AfroScope-Models trained on this data. It introduces Mirror-Serengeti for hierarchical disambiguation to resolve fine-grained confusions among related varieties and analyzes cross-lingual transfer and domain effects to guide robust deployment. The results show data-size thresholds, domain and script influences, and substantial gains from targeted disambiguation (average macro-F1 improvement of +4.55 on confusable groups). The resources and insights enable scalable measurement of Africa's linguistic landscape and offer practical guidance for building inclusive, robust African NLP systems, while noting limitations such as code-switching and metadata dependence that warrant future work.

Abstract

Language Identification (LID) is the task of determining the language of a given text and is a fundamental preprocessing step that affects the reliability of downstream NLP applications. While recent work has expanded LID coverage for African languages, existing approaches remain limited in (i) the number of supported languages and (ii) their ability to make fine-grained distinctions among closely related varieties. We introduce AfroScope, a unified framework for African LID that includes AfroScope-Data, a dataset covering 713 African languages, and AfroScope-Models, a suite of strong LID models with broad language coverage. To better distinguish highly confusable languages, we propose a hierarchical classification approach that leverages Mirror-Serengeti, a specialized embedding model targeting 29 closely related or geographically proximate languages. This approach improves macro F1 by 4.55 on this confusable subset compared to our best base model. Finally, we analyze cross linguistic transfer and domain effects, offering guidance for building robust African LID systems. We position African LID as an enabling technology for large scale measurement of Africas linguistic landscape in digital text and release AfroScope-Data and AfroScope-Models publicly.

Figures (6)

• Figure 1: The AfroScope framework begins with Dataset & Coverage, where we maximize language coverage by aggregating multilingual datasets and metadata (language families, scripts) to construct the AfroScope-Data. We also employ rigorous decontamination to produce high-quality Evaluation Data. In Model & Analysis, we fine-tune baseline architectures introducing AfroScope-Models. To address fine-grained distinctions between closely related languages, we introduce Mirror-Serengeti for hierarchical classification. Finally, we evaluate these components through extensive cross-lingual analysis.
• Figure 2: Distribution of languages across major language groupings, intermediate sub-families, and finer-grained groupings, capturing their genetic relationships.
• Figure 3: Relationship between training data size (log scale) and average macro-F$_1$ across low-resource, medium-resource, and high-resource languages.
• Figure 4: Per-language macro-F$1$ scores across domains. Bubble size corresponds to training examples.
• Figure 5: UMAP visualization comparing base Serengeti (top) and Mirror-Serengeti (bottom) embedding spaces. We visualize five groups representing macro-languages and confusion pairs. Specialized embeddings show improved separation between closely related language varieties.