# Multi-dialect support

This is Denmark's most advanced speech recognition model, trained by Alvenir as part of the CoRal project, based on the CoRal-v2 dataset, covering various Danish dialects.

Denmark's state-of-the-art speech recognition model trained by Alvenir, based on the CoRal-v2 dataset, supporting multiple Danish dialects

This is a version of the NLLB-1.3b model fine-tuned with parallel data for 29 Finno-Ugric languages, supporting the generation of multiple dialects/variants.

A phoneme recognition model fine-tuned based on wav2vec2-base, supporting English speech to International Phonetic Alphabet (IPA) conversion

Uzbek automatic speech recognition model fine-tuned from OpenAI Whisper Medium

A high-performance Arabic information retrieval model built on the sentence-transformers framework, optimized for the richness and complexity of the Arabic language.

A lightweight and accelerated version of the Norwegian automatic speech recognition model developed by the National Library of Norway, reducing parameter count through distillation while maintaining transcription quality.

A Cantonese and English mixed speech recognition model optimized based on the Whisper architecture, supporting high-precision bilingual transcription

This model is an automatic speech recognition (ASR) model fine-tuned on Burmese speech datasets based on openai/whisper-tiny, supporting Burmese speech-to-text tasks.

This is a transformers model for Arabic alphabet speech classification, capable of recognizing and classifying the pronunciation of Arabic letters.

Nepali male voice synthesis model based on VITS architecture, supporting high-quality text-to-speech functionality

A speech-based accent classifier for distinguishing Brazilian Portuguese from other accents.

This is a Shan language text-to-speech (TTS) model designed to convert Shan text into natural speech.

A translation model for Darija (Moroccan Arabic) to Modern Standard Arabic (MSA), trained on 26,000 manually annotated and GPT-4 enhanced text pairs

An automatic speech recognition model developed by the National Library of Norway, based on the OpenAI Whisper architecture, supporting transcription in Norwegian and English.

An automatic Norwegian speech recognition model launched by the National Library of Norway, developed based on OpenAI's Whisper architecture, supporting multiple Norwegian dialects and English.

Arabic speech recognition model based on Whisper-large-v3, optimized for Moroccan accent

Norwegian automatic speech recognition model developed based on OpenAI Whisper, with additional training for lowercase, punctuation-free verbatim transcription

An automatic speech recognition model developed by the National Library of Norway, based on the Whisper architecture, supporting speech transcription and translation of Norwegian and English.

Whisper base model fine-tuned on Malaysian datasets, supporting Malay and English speech recognition

NorBERT 3 xs is a BERT model optimized for Norwegian, the smallest version in the new generation NorBERT language model series with 15M parameters.

NorBERT 3 is a next-generation Norwegian language model based on the BERT architecture, supporting both Bokmål and Nynorsk written Norwegian.

This model is a fine-tuned version of OpenAI's Whisper Large-V2 for Hausa speech recognition tasks, trained on the Common Voice 11.0 dataset

Georgian automatic speech recognition model fine-tuned based on OpenAI Whisper-small

This is a fine-tuned version of the OpenAI Whisper Large V2 model for Malayalam speech recognition tasks, trained using the Common Voice 11.0 dataset

An acoustic model for Spanish automatic speech recognition, fine-tuned for 5 epochs based on facebook/wav2vec2-large-xlsr-53 using approximately 944 hours of Spanish data.

This model is an automatic speech recognition (ASR) model fine-tuned on the Norwegian Bokmål dialect speech dataset based on facebook/wav2vec2-xls-r-1b

This is a neural machine translation model for English to Portuguese (including Brazilian Portuguese), part of the OPUS-MT project.

Hindi automatic speech recognition (ASR) model based on Wav2Vec2 architecture, fine-tuned on public speech datasets

Automatic speech recognition model trained on a large-scale Arabic speech dataset

A Hindi automatic speech recognition model fine-tuned on low-resource Indian language datasets based on facebook/wav2vec2-large-xlsr-53

A Breton fine-tuned speech recognition model based on facebook/wav2vec2-large-xlsr-53

This is an automatic speech recognition model fine-tuned on the Urdu Common Voice 7 dataset based on facebook/wav2vec2-xls-r-300m.

This is an automatic speech recognition (ASR) model fine-tuned on the Hindi Common Voice 7.0 dataset based on facebook/wav2vec2-xls-r-300m

A speech recognition model fine-tuned on the Breton Common Voice dataset based on facebook/wav2vec2-large-xlsr-53

This is a Punjabi automatic speech recognition model fine-tuned on the Common Voice dataset based on Harveenchadha/vakyansh-wav2vec2-punjabi-pam-10

A Portuguese speech recognition model fine-tuned on the Common Voice dataset based on Edresson/wav2vec2-large-xlsr-coraa-portuguese

This model is an automatic speech recognition model fine-tuned on the Common Voice 8 Basaa dataset based on facebook/wav2vec2-xls-r-1b.

This is an automatic speech recognition (ASR) model fine-tuned on Assamese datasets based on the facebook/wav2vec2-xls-r-300m model

This model is an automatic speech recognition model fine-tuned on Tatar language datasets based on facebook/wav2vec2-xls-r-1b, achieving a word error rate (WER) of 16.87% on the Common Voice 8 dataset.