# Low-resource language processing

A Seraiki news summarization model fine-tuned based on the mBART-50 multilingual model, capable of generating concise summaries from Seraiki news content

This is a speech recognition model fine-tuned on the Common Voice dataset based on facebook/wav2vec2-xls-r-300m, supporting Irish language.

The Camel Model is a text generation model based on the transformer architecture, supporting Azerbaijani and trained using reinforcement learning.

This model is a fine-tuned version of OpenAI's Whisper Small on the FLEURS dataset, focusing on speech recognition tasks and supporting Telugu (te).

A text summarization model fine-tuned on Sinhala news data based on Google's mT5-small model

XLM-RoBERTa-large fine-tuned Uzbek named entity recognition model supporting 21 entity types

A speech recognition model fine-tuned on the Polish Common Voice 17.0 dataset based on OpenAI Whisper-base

Arabic-to-English translation model fine-tuned on the kde4 dataset based on Helsinki-NLP/opus-mt-ar-en

This model is fine-tuned from google/mt5-small for converting Romanized Nepali to Nepali text

This is a Transformer-based speech model supporting Dhivehi (Maldivian) speech processing tasks.

A headline generation model fine-tuned on Ukrainian news datasets based on the multilingual mT5 model, capable of generating concise and accurate headlines for Ukrainian news articles.

A Sinhala speech recognition model fine-tuned based on openai/whisper-small, supporting conversion of Sinhala speech to text.

Kyrgyz automatic speech recognition (ASR) model based on the Whisper architecture, developed with support from the National Commission on Language and Language Policy under the President of the Kyrgyz Republic

KuBERT is a Central Kurdish language model based on the BERT framework, designed to address the scarcity of Kurdish language resources and enhance computational linguistics capabilities.

A fine-tuned Amharic text summarization model based on google/mt5-small, suitable for news article headline generation tasks.

A Polish sentence embedding model based on RoBERTa architecture, focusing on sentence similarity calculation and feature extraction tasks.

NLLB-CLIP-SigLIP is a multilingual vision-language model that combines the text encoder from NLLB and the image encoder from SigLIP, supporting 201 languages.

A translation model fine-tuned on Russian-Kabardian datasets based on facebook/m2m100_1.2B

This is a sentence-transformers based model for mapping Sinhala sentences into a 768-dimensional vector space, supporting tasks like sentence similarity calculation and semantic search.

This is a Spanish-to-Portuguese translation model fine-tuned based on the T5-small architecture, trained on the tatoeba dataset, with an evaluated BLEU score of 11.2994.

This is a Bodo language configuration model based on the RoBERTa architecture, including a byte-level BPE tokenizer for Bodo and RoBERTa base configuration.

This model is a fine-tuned version of whisper-small-cv11-french, optimized for Haitian Creole speech recognition

This is a Transformer model for Turkish text punctuation restoration, capable of predicting the correct positions of periods (.), commas (,), and question marks (?).

Glot500 is a multilingual pre-trained model that supports over 500 languages and is trained based on the masked language modeling (MLM) objective.

Dutch BERT-based sentence embedding model that maps text to a 768-dimensional vector space, suitable for semantic similarity calculation and text classification tasks

Hungarian abstractive summarization model trained on the mT5-small architecture using the HunSum-1 dataset

This model is a fine-tuned version of openai/whisper-small on the google/fleurs yo_ng dataset, designed for automatic speech recognition tasks in Yoruba.

Slovak speech recognition model fine-tuned on OpenAI Whisper-small, trained on the Common Voice 11.0 Slovak dataset

This model is a Q&A model fine-tuned on the Slovak language dataset skquad based on SlovakBERT

A language model pretrained on large-scale Bulgarian and Macedonian texts, part of the MaCoCu project

Estonian feature extraction model fine-tuned based on XLM-RoBERTa base model

A multilingual processing model supporting over 100 languages and writing systems, covering major global language families and dialect variants

This model is an English-to-Arabic translation model fine-tuned on the kde4 dataset based on Helsinki-NLP/opus-mt-en-ar.

A generative summarization model fine-tuned on Persian summarization datasets based on MBART-large-50

A Persian summarization model fine-tuned on pn_summary dataset based on google/mt5-base

A multilingual summarization model based on mT5, specifically fine-tuned for Arabic on the XLSum dataset

This model is a fine-tuned summarization model based on google/mt5-base on the Urdu xlsum dataset

Urdu speech recognition model trained on Common Voice dataset, using wav2vec 2.0 architecture

A text summarization model fine-tuned on the Arabic summarization dataset xlsum based on the AraBART model

This model is used to predict punctuation marks in Catalan, capable of restoring periods, commas, question marks, hyphens, and colons.