logo

Hive AI

Optical Character Recognition Model

Optical Character Recognition Model

Extract overlay, scene, and document text in 15+ languages, plus emojis, from images and video

About Hive’s OCR Model

Hive's Optical Character Recognition model detects and transcribes each word in an image. It also returns semantically grouped and ordered text blocks (Block Text) in their natural reading order for detections that are spatially close.


Our OCR model also supports emoji detection and classification for Apple, Samsung, and Google devices (over 3,000 emojis for each).


For documents that have a structure, such as receipts, Hive provides APIs that interpret the structure — such as returning each item on a receipt, its unit price, quantity, and total price


To learn about our OCR Moderation model, please see the OCR Moderation page

Hive's Optical Character Recognition model detects and transcribes each word in an image. It also returns semantically grouped and ordered text blocks (Block Text) in their natural reading order for detections that are spatially close.


Our OCR model also supports emoji detection and classification for Apple, Samsung, and Google devices (over 3,000 emojis for each).


For documents that have a structure, such as receipts, Hive provides APIs that interpret the structure — such as returning each item on a receipt, its unit price, quantity, and total price


To learn about our OCR Moderation model, please see the OCR Moderation page

Comprehensive coverage for diverse use cases

Comprehensive coverage for diverse use cases

Our deep learning model accurately transcribes text and emojis from image-based text across key domains and languages.

Input
Response Fields
Domains
Emojis

Language Support

Arabic

Arabic

Bengali

Bengali

Chinese

Chinese

Dutch

Dutch

English

English

French

French

German

German

Gujrati

Gujrati

Hindi

Hindi

Indonesian

Indonesian

Italian

Italian

Japanese

Japanese

Korean

Korean

Marathi

Marathi

Malay

Malay

Norwegian

Norwegian

Persian

Persian

Polish

Polish

Portuguese

Portuguese

Romanian

Romanian

Russian

Russian

Spanish

Spanish

Tamil

Tamil

Tagalog

Tagalog

Telugu

Telugu

Turkish

Turkish

Vietnamese

Vietnamese

Emoji

Emoji

See our OCR Model in action

Note: This model is optimized for images with 150 words or fewer. For images with more words than that, we recommend splitting the image into multiple segments and submitting them separately. Use of this demo is subject to our site’s Terms of Service.

Simple usage based pricing so you only pay for what you use

Optical Character Recognition Model Pricing Details

Model
Unit

OCR (Image)

$1.50

1000 Requests

OCR (Video)

$0.10

Minute

How customers use our OCR Model

Why choose our OCR Model

Why choose our OCR Model

Speed at scale

Speed at scale

We handle high volume with ease and efficiency, serving real-time responses to billions of API calls per month.
Simple integration

Simple integration

Model predictions are accessible with a single API call. Integrate our OCR model into any application with just a few lines of code.
Proactive updates

Proactive updates

Our OCR model is regularly upgraded to improve performance, add commonly requested features, and keep up with evolving customer needs.

Ready to build something?

AI Models

Applications

Platform Solutions

Media Solutions

Company

Other Site Pages

Contact Us

footer-hive-logo
© Copyright 2024