Skip to content

Convert captured images to text using BaiduOCR, GoogleOCR, WindowsOCR, tesseractOCR, RapidOCR or Capture2Text, and translate the resulting text using Google, Chatgpt, Edgegpt, DeepL or many more. Desktop application with a nice GUI provided by customtkinter.

License

Notifications You must be signed in to change notification settings

Azornes/ocrTranslator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

81 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

OCR Translator

Convert captured images into text and then translate that text.

Downloads python 3.9


With this app, you can select your preferred OCR and translation services. After clicking on START or using the keyboard shortcut Alt+Win+T, the program will launch and you can choose the area of the screen to scan for text using OCR. If you have selected a translation service, the text will then be automatically translated.

preview:

show.preview.OCR.Translator.mp4

πŸ”₯ Features

  • Desktop application with a user-friendly graphical user interface (GUI) provided by customtkinter.
  • Ability to select preferred OCR and translation services.
  • Option to run the program using either the START button or the keyboard shortcut (Alt+Win+T or bound from options).
  • Capability to choose the area of the screen to scan for text using OCR and save the position (for example, when watching a movie and the subtitles always appear in one spot, so you don't have to select the text area again).
  • Automatic translation of the captured text if a translation service has been selected.
  • Ability to capture subtitles from movies or games by selecting the corresponding area of the screen and displaying the translated text next to them.
  • Chat with chatGPT or edgeGPT.
  • Ability to translate from the clipboard or manually entered text (similar to a typical translation app).
  • Save all selected options and settings to a file and load them when the program is launched.

Desktop App

Download the desktop app here Tested only on Windows 10.


Dependency

  1. Python 3.9. (If you want run from source)
  2. (optional) Capture2Text.
  3. (optional) Tesseract.
  4. (optional) Google api generate a service_account_creds.json. Then, put file into the ocrTranslate/configs directory.

5. (optional) ChatGPT

Configuration

  1. Create account on OpenAI's ChatGPT
  2. Save your email and password

Authentication method: (Choose 1 and paste to app settings)

- Email/Password

Currently broken for free users. Do export PUID="..." if you have a plus account. The PUID is a cookie named _puid Not supported for Google/Microsoft accounts.

- Access token

https://chat.openai.com/api/auth/session

6. (optional) EdgeGPT

Checking access (Required)

  • Install the latest version of Microsoft Edge
  • Alternatively, you can use any browser and set the user-agent to look like you're using Edge (e.g., Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/111.0.0.0 Safari/537.36 Edg/111.0.1661.51). You can do this easily with an extension like "User-Agent Switcher and Manager" for Chrome and Firefox.
  • Open bing.com/chat
  • If you see a chat feature, you are good to go

Getting authentication (Required)

  • Install the cookie editor extension for Chrome or Firefox
  • Go to bing.com
  • Open the extension
  • Click "Export" on the bottom right, then "Export as JSON" (This saves your cookies to clipboard)
  • Paste your cookies into a file cookies.json
  • Paste your file cookies.json to ocrTranslate/configs/

πŸ“Š Tables with information

Supported OCR Services

ID OCR Internet/Local Status
1 Google Vision Api Internet stable
2 Google Vision Free Demo Internet stable
3 Baidu Api Internet stable
4 Windows OCR Local stable
5 Capture2Text Local stable
6 Tesseract Local stable
7 RapidOCR Local stable

Supported Translation Services

ID Translator Number of Supported Languages Advantage Service Status
1 Niutrans 302 support the most languages in the world Northeastern University / Niutrans, China /
2 Alibaba 221 support most languages, support professional field Alibaba, China stable
3 Baidu 201 support most languages, support professional field, support Classical Chinese Baidu, China stable
4 Iciba 187 support the most languages in the world Kingsoft / Xiaomi, China stable
5 MyMemory 151 support the most languages in the world, good at Creole English, Creole French Translated, Italy stable
6 Iflytek 140 support the most languages in the world Iflytek, China /
7 Google 134 support more languages in the world Google, America stable(offline in China inland)
8 VolcEngine 122 support more languages in the world, support professional field ByteDance, China /
9 Lingvanex 112 support translation of different regions but the same language Lingvanex, Cyprus stable
10 Bing 110 support more languages in the world Microsoft, America stable
11 Yandex 102 support more languages in the world, support word to emoji Yandex, Russia /
12 Itranslate 101 support translation of different regions but the same language, such as en-US, en-UK, en-AU Itranslate, Austria stable
13 Sogou 61 support more languages in the world Tencent, China stable
14 ModernMt 56 open-source, support more languages in the world Modernmt / Translated, Italy stable
15 SysTran 52 support more languages in the world SysTran, France stable
16 Apertium 45 open-source Apertium stable
17 Reverso 42 popular on Mac and Iphone Reverso, France stable
18 CloudYi 28 support main languages Xiamen University / CloudTranslation, China stable
19 Deepl 27 high quality to translate but response slowly Deepl, Germany stable
20 QQTranSmart 22 support main languages Tencent, China stable
21 TranslateCom 21 good at English translation TranslateCom, America stable
22 Tilde 21 good at lv, de, fr translation Tilde, Latvia /
23 QQFanyi 17 support main languages Tencent, China stable
24 Argos 17 open-source Argos / Libre, America stable
25 TranslateMe 16 good at English translation TranslateMe / Neosus, Lithuania stable
26 Youdao 15 support main languages, high quality Netease, China stable
27 Papago 15 good at Korean translation Naver, South Korea stable
28 Marai 15 good at Japanese translation MaraiTranslate, Japan /
29 Iflyrec 12 good at Chinese translation Iflytek, China stable
30 Yeekit 10 support main languages CTC, China stable
31 LanguageWire 8 good at English translation LanguageWire, Denmark stable
32 Caiyun 7 high quality to translate but response slowly, support professional field ColorfulClouds, China stable
33 Elia 6 good at Basque translation Elhuyar, Spain stable
34 Judic 4 good at European translation CrossLang, Belgium stable
35 Mglip 3 good at Mongolia translation Inner Mongolia University, China stable
36 Utibet 2 good at Tibet translation Tibet University, China stable

About

Convert captured images to text using BaiduOCR, GoogleOCR, WindowsOCR, tesseractOCR, RapidOCR or Capture2Text, and translate the resulting text using Google, Chatgpt, Edgegpt, DeepL or many more. Desktop application with a nice GUI provided by customtkinter.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published