# African language support

A text-to-speech synthesis model developed for the Bambara language, based on VITS architecture and Facebook MMS foundation model

A Hausa speech recognition model fine-tuned from facebook/w2v-bert-2.0, trained on 500 hours of Hausa data with a word error rate of 7.47%

This is a Bambara automatic speech recognition model fine-tuned based on the NVIDIA NeMo framework, suitable for Bambara speech-to-text tasks.

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53, specifically designed for automatic speech recognition tasks in Swahili.

This model is a fine-tuned machine translation model for Kinyarwanda to English, based on Helsinki-NLP/opus-mt-rw-en

This model is an automatic speech recognition (ASR) model fine-tuned on Amharic speech corpus based on facebook/wav2vec2-large-xlsr-53.

Luganda text-to-speech system based on Tacotron2 architecture, trained with professional studio data

BigScience Large Open-science Multilingual Language Model, supporting 46 natural languages and 12 programming languages, developed through international collaboration

This is an automatic speech recognition model for Wolof, based on the wav2vec 2.0 architecture, trained on the DVoice dataset, supporting Wolof speech transcription.

BigScience Large Open-science Multilingual Language Model, supporting 46 natural languages and 12 programming languages

An Amharic language model trained based on the Roberta architecture, suitable for various natural language processing tasks.

This is a masked language model focused on the Tswana language, aiming to enhance the performance of low-resource South African languages in the field of NLP.

Swahili automatic speech recognition model fine-tuned on XLSR-53 large model, supports 16kHz sampling rate audio input

This is a machine translation model based on the Transformer architecture, specifically designed for translating Finnish to the Congo dialect of Swahili.

Transformer-based machine translation model supporting Kuanyama (kj) to English (en) translation tasks

A Transformer-based machine translation model supporting translation tasks from Tumbuka (tum) to English (en), released by the Helsinki-NLP team.

This is a Transformer-based multilingual machine translation model for Bantu languages to English, supporting translation tasks for 12 Bantu languages.

This is a customized and distilled version of bert-base-multilingual-cased, specifically designed for Swahili language processing, capable of fully reproducing the representations generated by the original model while maintaining original accuracy.

OPUS-MT translation model supporting machine translation tasks from Twi (tw) to Swedish (sv).

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

A machine translation model fine-tuned based on facebook/m2m100_418M, achieving strong baseline performance for English to Yoruba text translation