# Multi-dataset fusion

A compact version fine-tuned based on F5-TTS, trained with 150 hours of Vietnamese speech data, for research purposes only.

A BERT-base fine-tuned model for paraphrase detection, suitable for duplicate content identification, Q&A systems, and semantic similarity analysis.

An automatic speech recognition model fine-tuned for Hebrew based on OpenAI Whisper Large v3 Turbo

voc2vec is a foundational model specifically designed for non-linguistic human data, built on the wav2vec 2.0 framework, with a pretraining dataset covering approximately 125 hours of non-linguistic audio.

This is a large-scale Uzbek speech recognition model based on the FastConformer architecture, supporting both Transducer and CTC decoding, and demonstrating excellent performance across multiple test sets.

A reward model trained on google/gemma-2b-it for evaluating text generation quality

orca_mini_3b is a text generation model trained on the OpenLLaMa-3B model, utilizing instructions and inputs from the WizardLM, Alpaca, and Dolly-V2 datasets for explanatory fine-tuning, and applying the dataset construction methods from the Orca research paper.

A text-to-image generation model based on stable diffusion technology, specializing in high-quality anime-style images.

Speech recognition model optimized for Brazilian Portuguese, trained with 453 hours of audio from 7 public datasets

A multi-dataset pre-trained natural language inference model based on the RoBERTa-Large architecture, integrating multiple well-known NLI datasets including SNLI, MNLI, FEVER-NLI, and ANLI.