Recent progress in Automatic Speech Recognition (ASR) has primarily benefited high-resource standard languages, while dialectal speech remains challenging and underexplored. We present an expanded benchmark for low-resource Modern Greek dialects, covering Aperathiot, Cretan, Lesbian, and Cappadocian, spanning southern, northern, and contact-influenced varieties with varying degrees of divergence from Standard Modern Greek. The benchmark provides dialectal transcriptions in the Greek alphabet, following SMG-based orthographic conventions, while preserving dialectal lexical and morphophonological forms. Using this benchmark, we evaluate state-of-the-art multilingual ASR models in a zero-shot setting and by further fine-tuning per dialect. Zero-shot results reveal a clear performance gradient with dialectal distance from Standard Modern Greek, with best WERs ranging from about 60-70% for southern dialects to over 80% for Lesbian and nearly 97% for Cappadocian. Fine-tuning substantially reduces error rates (up to 47% relative WER improvement), with Cappadocian remaining the most challenging variety (best WER 68.17%). Overall, our results highlight persistent limitations of current pretrained ASR models under dialectal variation and the need for dedicated benchmarks and adaptation strategies.

This paper presents the first treebank for the dialect of Lesbos, a low-resource living Northern variety of Modern Greek (MG), annotated according to the Universal Dependencies (UD) framework. So far, the only dialectal treebank available for Greek developed with cross-dialectal knowledge transfer is an East Cretan one, which belongs to the same Southern branch as Standard Modern Greek (SMG). Our study investigates the effectiveness of cross-dialectal knowledge transfer between dialectologically less similar varieties of the same language by leveraging knowledge from SMG to annotate the Northern dialect of Lesbos. We describe the annotation process, present the resulting treebank, inject additional linguistic knowledge to enhance the results, and evaluate the effectiveness of cross-dialectal knowledge transfer for active annotation. Our findings contribute to a better understanding of how dialectal variation within language families affects knowledge transfer in the UD framework, with implications for other low-resource varieties.