T

Trocr Base Spanish

Developed by qantev
Base version of TrOCR model, specifically designed for Spanish printed text, based on Transformer architecture, fine-tuned on a custom dataset
Downloads 170
Release Time : 2/22/2024

Model Overview

Transformer-based optical character recognition model for converting printed text images to text, does not support handwriting recognition

Model Features

Spanish Optimization
Fine-tuned on a custom dataset of 2 million Spanish samples, specifically optimized for Spanish OCR performance
Dynamic Image Generation
Utilizes dynamic image generation technology during training, more efficient compared to pre-stored images
Printed Text Specialization
Specifically designed for printed text, does not support handwriting recognition

Model Capabilities

Printed text image to text conversion
Spanish OCR
Short text recognition (up to 10 words)

Use Cases

Document Digitization
Wikipedia Content Extraction
Extract text content from Spanish Wikipedia page images
Form Processing
XFUND Dataset Processing
Process form images in the Spanish XFUND dataset
CER 0.0732 / WER 0.2028
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
Š 2025AIbase