# Learning the Arabic Dialect Continuum as a Continuous Space: A Regression Approach to Speaker Origin Prediction

arXiv: https://arxiv.org/abs/2607.19751 · HTML: https://mlsanity.com/paper/2607.19751/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 25 references that are in the map.
- First posted on arXiv 22 Jul 2026.

## Builds on (most cited first, 12 of 12)

- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://mlsanity.com/paper/2111.09296/) (2021)
- [AraBERT: Transformer-based Model for Arabic Language Understanding](https://mlsanity.com/paper/2003.00104/) (2020)
- [GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization](https://mlsanity.com/paper/2309.16020/) (2023)
- [ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic](https://mlsanity.com/paper/2101.01785/) (2020)
- [Arabic Dialect Identification in the Wild](https://mlsanity.com/paper/2005.06557/) (2020)
- [ALDi: Quantifying the Arabic Level of Dialectness of Text](https://mlsanity.com/paper/2310.13747/) (2023)
- [NADI 2024: The Fifth Nuanced Arabic Dialect Identification Shared Task](https://mlsanity.com/paper/2407.04910/) (2024)
- [Automatic Dialect Detection in Arabic Broadcast Speech](https://mlsanity.com/paper/1509.06928/) (2015)
- [Multi-view Dimensionality Reduction for Dialect Identification of Arabic Broadcast Speech](https://mlsanity.com/paper/1609.05650/) (2016)
- [The Arabic Generality Score: Another Dimension of Modeling Arabic Dialectness](https://mlsanity.com/paper/2508.17347/) (2025)
- [ARCADE: A City-Scale Corpus for Fine-Grained Arabic Dialect Tagging](https://mlsanity.com/paper/2601.02209/) (2026)

## Built on by (most cited first, 1 of 1)

- [Do Speech Representations Preserve Regional Accent Across Read and Spontaneous Speech?](https://mlsanity.com/paper/2610.06430/) (2026)

## Abstract (arXiv)

We present a regression-based approach to Arabic dialect geolocation that models dialectal variation as a continuous geographic space rather than discrete categories. Speaker origin is predicted as continuous latitude-longitude coordinates using a hierarchical neural architecture that fuses frame-level XLS-R-300M and Whisper-large-v3 encoder representations with phonotactic descriptors through a Transformer encoder and a learnable attention-pooled query. A spherical geodesic loss directly optimizes great-circle distance on Earth's surface, avoiding distortions inherent to planar coordinate regression. Under a leakage-free 5-fold GroupKFold protocol grouped by source recording, our model attains a pooled median localization error of 481.2 km. Auxiliary country and city heads reach 64.5% and 45.2% accuracy, respectively. A permutation Mantel test on the learned latent space provides quantitative support for the Arabic dialect continuum hypothesis. To probe true generalization, we further introduce a city-masking protocol in which two cities per fold are removed from training but retained in validation. Under this zero-shot regime, the mean error rises to 1173.3 km, a 1.32x degradation relative to seen cities. Our findings establish continuous geographic modeling as a principled framework for Arabic dialect geolocation and quantify both its strengths and the substantial headroom that remains.