# Multi-dataset Training

This is an image-to-image conversion model based on LoRA technology, primarily used for non-commercial image editing tasks.

ViTPose++ is a vision Transformer-based foundation model for human pose estimation, achieving an outstanding performance of 81.1 AP on the MS COCO keypoint test set.

kazRush-ru-kk is a Russian-to-Kazakh translation model based on the T5 configuration, trained on multiple open-source parallel datasets.

RAD-DINO-MAIRA-2 is a vision transformer model trained with DINOv2 self-supervised learning, specifically designed for encoding chest X-ray images.

This model, fine-tuned from Wav2Vec2-Large-Robust, can predict the speaker's age and gender from raw audio.

This is a Bengali text summarization model based on the GPT2 architecture, specifically optimized for news content.

This model is fine-tuned on the Common Voice, JVS, and JSUT datasets for Japanese speech recognition tasks using openai/whisper-base.

This is a German automatic speech recognition model based on the FastConformer architecture, employing a hybrid training approach with Transformer and CTC, with a parameter size of approximately 115M.

A Korean text summarization model based on the T5 architecture, specifically optimized for Korean text to generate concise and accurate summaries.

This is a named entity recognition model fine-tuned on PubMedBERT, specifically designed to identify gene and protein entities in biomedical texts.

This is a sentence embedding model based on the MiniLM architecture, capable of mapping text to a 384-dimensional vector space, suitable for tasks such as semantic search and sentence similarity calculation.

A Cantonese speech recognition model fine-tuned based on OpenAI Whisper-small, achieving a CER of 7.93 on the Common Voice 16.0 test set

This is a large Conformer-Transducer model for Spanish automatic speech recognition, with approximately 120 million parameters, trained on 1340 hours of Spanish speech data.

This is a large Conformer-CTC model for Spanish automatic speech recognition (ASR), trained and released by NVIDIA.

This is a large-scale Conformer-Transducer model for French automatic speech recognition, with approximately 120 million parameters, trained on over 1,500 hours of French speech data.

This is a large-scale Conformer-CTC model for German automatic speech recognition, trained and optimized by NVIDIA on thousands of hours of German speech data.

NVIDIA Streaming Citrinet 1024 is a non-autoregressive model for English automatic speech recognition, based on CTC loss/decoding, with approximately 140 million parameters.

A pre-trained model based on the DenseNet121 architecture, specifically designed for chest X-ray image classification tasks with 18 output targets.

This is a sentence embedding model based on sentence-transformers, capable of mapping text to a 384-dimensional vector space, suitable for semantic search and clustering tasks.

Optimized automatic speech recognition model for Galician, fine-tuned based on wav2vec2-large-xlsr-53, with a WER of 7.12

This is a Wav2vec 2.0 speech recognition model optimized for Brazilian Portuguese, fine-tuned on multiple Brazilian Portuguese datasets

Japanese speech recognition model fine-tuned from facebook/wav2vec2-large-xlsr-53, supporting 16kHz sampling rate audio input

This is a German automatic speech recognition model based on the XLS-R 1B architecture, fine-tuned on multiple German speech datasets including Common Voice 8.0

A bilingual summarization model based on multilingual T5, supporting English and German text summarization tasks

A Wav2vec 2.0 speech recognition model fine-tuned on Brazilian Portuguese datasets, supporting automatic speech recognition tasks for Brazilian Portuguese.

This model is a Wav2Vec2 speech recognition model fine-tuned on the Common Voice Turkish dataset, demonstrating excellent performance in Turkish automatic speech recognition tasks.

A sentence transformation model based on RoBERTa-large, specifically designed for sentence similarity tasks, trained on ANLI, MNLI, and SNLI datasets

A Japanese automatic speech recognition model fine-tuned from facebook/wav2vec2-large-xlsr-53 using the Common Voice and JSUT datasets.

A Chinese speech recognition model fine-tuned based on facebook/wav2vec2-large-xlsr-53, supporting 16kHz sampling rate audio input.

This is a Wav2vec 2.0 model fine-tuned for Brazilian Portuguese, trained using multiple open Brazilian Portuguese datasets including Common Voice, MLS, CETUC, etc.

A Japanese speech recognition model fine-tuned based on facebook/wav2vec2-large-xlsr-53, supporting hiragana output

This is a Spanish automatic speech recognition model fine-tuned based on the XLS-R 1 billion parameter model, trained and optimized on multiple Spanish datasets.

This model is fine-tuned from CamemBERT-base, specifically designed for answer extraction tasks in French texts, trained on SquadFR, FQuAD, and PIAF datasets.

A model specialized in Thai text sentiment analysis, supporting positive, neutral, and negative sentiment classification.

A DistilBERT model fine-tuned for Persian Named Entity Recognition (NER) tasks, supporting recognition of 10 entity types.

A binary natural language inference model trained on 8 NLI datasets, excelling in long-text reasoning tasks

Japanese speech recognition model fine-tuned from facebook/wav2vec2-large-xlsr-53, supports 16kHz sampling rate input