# Low-Resource Language Support

A translation model fine-tuned on mbart-large-50-many-to-many-mmt for English to Bhojpuri text translation

YarnGPT Local Version is a text-to-speech model specifically designed for Yoruba, Igbo, and Hausa languages, utilizing pure language modeling technology to deliver high-quality, natural, and culturally appropriate speech synthesis.

A VITS-based text-to-speech model supporting 13 Indian languages with diverse speaking styles and emotional expressions

This model is a fine-tuned English-Somali translation model based on facebook/nllb-200-distilled-600M, achieving a BLEU score of 38.1 on the evaluation set.

A Telugu automatic speech recognition model based on OpenAI Whisper-small architecture, optimized for Indian languages by the ARTPARK-IISc team

Teuken-7B-instruct-commercial-v0.4 is an instruction-tuned 7-billion-parameter multilingual large language model supporting all 24 official EU languages, specifically designed for commercial and research scenarios.

This is a 🤗 transformers model card for audio-to-text tasks, which has been pushed to the Hub platform.

This is a Tacotron2-based text-to-speech (TTS) system trained on the Luganda CommonVoice dataset, specifically optimized for female voices.

Typhoon-Audio Preview is a Thai and English audio-language model capable of processing text and audio inputs, with text outputs.

AfroLlama3 is an 8-billion-parameter language model optimized from Meta-Llama-3-8B, specifically enhanced for text processing and generation in Swahili, English, Zulu, Xhosa, Yoruba, and Hausa.

A text-to-speech system developed for Central Kurdish (ckb), capable of converting written text into natural and fluent speech output.

A Shan text-to-speech model based on the MIT license, supporting the conversion of Shan text to speech.

SeaLMMM-7B is a multilingual multimodal model for Southeast Asia, supporting various Southeast Asian languages and visual tasks.

Honyaku-7b-v2 is an improved version of its predecessor, with enhanced accuracy in multilingual tag adherence.

The SambaLingo-Turkish-Chat Model is a human preference-aligned chat model supporting Turkish and English, adapted from Llama-2-7b with Turkish language adaptation and direct preference optimization training.

This is a multilingual semantic model trained using knowledge distillation methods, specifically optimized for the Yoruba language, capable of handling cross-language sentence similarity tasks.

Kekchi text-to-speech model developed by Meta, part of the Massively Multilingual Speech project

Chichewa text-to-speech model developed by Meta AI, based on VITS architecture, supporting high-quality speech synthesis

An Ese Ehue text-to-speech model developed by Meta as part of the Massively Multilingual Speech project, supporting high-quality speech synthesis.

A Central Tibetan text-to-speech model developed by Meta, based on the VITS architecture, supporting high-quality speech synthesis

This is a Thai text-to-speech model based on the Tacotron2 architecture, trained using a modified Common Voice Thai dataset, with processed speech that does not retain original speaker characteristics.

This is a fine-tuned model for speech language identification (LID) across 512 languages, based on the Wav2Vec2 architecture, capable of recognizing the language category of input audio.

This is a speech language identification model based on the Wav2Vec2 architecture, capable of recognizing 256 languages, and is part of Facebook's Massively Multilingual Speech (MMS) project.

A language identification model fine-tuned from Facebook's Massively Multilingual Speech project, supporting audio classification for 126 languages

NLLB is a multilingual machine translation model supporting 200 languages, designed to cover the majority of global languages, especially low-resource languages.

Varta-T5 is a multilingual model pre-trained on the Varta dataset, supporting 14 Indian languages and English, primarily used for text generation and classification tasks.

The Telugu (te) natural language processing model provided by Stanza supports various linguistic analysis tasks from raw text to syntax analysis and entity recognition.

A Tamil pre-trained model based on the ELECTRA framework, with the second version trained on TPUs and expanded corpus scale

This model is a fine-tuned automatic speech recognition (ASR) model based on facebook/wav2vec2-large-xlsr-53 on an unknown dataset, supporting Iloko language speech recognition.

This is a pretrained language model for Tagalog, trained on multisource data, suitable for natural language processing tasks.

This is an English-to-Armenian machine translation model based on the transformer-align architecture, developed by the Helsinki-NLP team.