low-resource-languages
Filtering by topic low-resource-languages(3)Clear all filters
- PaperarXiv — Language & NLP (cs.CL)5 Jul 2026
Towards Digital Preservation of Efik: TTS for a Low-Resource African Language
Offiong Bassey Edet, Emmanuel Oyo-Ita, Archibong Okon Archibong, David Effanga Bassey et al.
This study presents the first end-to-end text-to-speech system for Efik, a low-resource tonal language in Nigeria. Four neural models were evaluated, with MMS-TTS achieving the highest MOS of 3.80, though tonal errors remained. The results provide a baseline and underscore the need for larger corpora and tone-aware modeling for tonal African languages.
Original abstract
Efik, a tonal language spoken by about 3 million second language speakers and 1.5 million native speakers in Southeastern Nigeria, remains underrepresented in speech synthesis research. We present the first documented end-to-end text-to-speech study for Efik, introducing a curated single speaker corpus of 2,632 utterances totaling three hours and a comparative evaluation of four neural models (VITS, MMS-TTS, SpeechT5, and Orpheus-TTS) under low resource conditions. Native speakers evaluated the systems using MOS, Nat-MOS, and A-MOS. MMS-TTS achieved the highest MOS of 3.80 +/- 0.63 and produced more stable long form speech, though tonal errors persisted. Other models showed greater tonal and prosodic inconsistencies. These results provide a reproducible baseline and highlight the need for larger corpora and tone aware modeling for tonal African languages.
- PaperarXiv — Language & NLP (cs.CL)4 Jun 2026
Reinforcement Learning Elicits Contextual Learning of Unseen Language Translation
Hanxu Hu, Zdeněk Šnajdr, Pinzhen Chen, Jannis Vamvas et al.
Large language models can be trained with reinforcement learning to learn to translate unseen languages by using in-context linguistic information, achieving better translations than fine-tuning or in-context learning alone.
Original abstract
Prior work has shown that large language models (LLMs) can translate unseen or low-resource languages by undergoing continued training or even by encoding a grammar book in their context. However, both methods typically overfit specific languages, with limited zero-shot transfer at test time. To translate extremely low-resource languages at scale, we argue that LLMs must acquire the meta-skill of utilizing in-context linguistic knowledge rather than memorizing specific languages. In this paper, we propose a reinforcement learning (RL) approach to unseen language translation given rich linguistic context, using a surface-level translation metric (chrF) as the reward. Empirically, despite the lightweight reward, our RL-trained models effectively extract and apply relevant linguistic information from the provided context, leading to better translations on completely unseen languages than in-context learning or supervised fine-tuning. Our analyses suggest that outcome-based RL can extend beyond conventional reasoning tasks like math and coding to serve as a recipe for language learning from context.
- PaperarXiv — Language & NLP (cs.CL)21 May 2026
From Lexicon to AI: A Structured-Data Pipeline for Specialized Conversational Systems in Low-Resource Languages
Siddhant Hitesh Mantri, Dhara Gorasiya, Malhar Kulkarni, Pushpak Bhattacharya
Transformed Hindi WordNet into 1.25 million instruction-response pairs to fine-tune a 12B-parameter language model using LoRA and 4-bit quantization. The resulting Hindi language learning chatbot achieved superior pedagogical effectiveness (91.0) compared to general-purpose models, demonstrating that structured lexical databases can effectively support conversational AI in low-resource languages.
Original abstract
Low-resource languages face a critical challenge in AI development: creating specialized conversational systems without access to massive training corpora. We present a systematic methodology for transforming structured linguistic resources into specialized AI systems, demonstrating that expert-curated lexical databases can serve as effective foundations for conversational AI development. Our approach converts Hindi WordNet into 1.25 million diverse instruction-response pairs, fine-tunes a 12B-parameter language model using resource-efficient LoRA with 4-bit quantization. Evaluation through a Hindi language learning chatbot demonstrates that structured-knowledge-based systems achieve superior pedagogical effectiveness (91.0 vs. 79.4-83.6 for general-purpose models) while maintaining competitive semantic performance and exceptional consistency. The complete pipeline demonstrates a proof-of-concept methodology using Hindi for developing specialized AI systems for any languages with WordNet resources. This work addresses the critical gap in AI accessibility for low-resource languages, offering a practical alternative to corpus-intensive approaches and potentially enabling specialized AI development for the hundreds of languages with existing WordNet resources.