I open sourced an all-purpose translation program on GitHub, github address: https://github.com/tianclll/Ace-Translate
Ace-Translate
A local offline translation program
Supports multiple translation scenarios
- Chinese to English and English to Chinese
- text translation
- word translation
- screenshot translation
- video translation
- Document translation, including TXT files, Excel, PPT, PDF, pictures and Word
- Document image translation
Install
Recommended to use python3.8
++paddlepaddle2.4.0
torch2.0.1
1. Pull code
git clone https://github.com/tianclll/Ace-Translate.git
cd Ace-Translate
2.Installation
2.1 Install PaddlePaddle
-
GPU
python3 -m pip install paddlepaddle-gpu==2.4.0 -i https://mirror.baidu.com/pypi/simple
-
CPU
python3 -m pip install paddlepaddle==2.4.0 -i https://mirror.baidu.com/pypi/simple
2.2 Install dependencies
pip install -r requirements.txt
- Windows requires additional downloads
pip install transformers[sentencepiece] pip install clip
2.3 Download model file
Click here to download
After unzipping, place it in the project folder (Ace-Translate).
2.4 Install Pyaudio
Need to run voice translation before installation
-
Linux
sudo apt-get install libasound2-dev wget https://files.portaudio.com/archives/pa_stable_v190700_20210406.tgz tar -xvf pa_stable_v190700_20210406.tgz cd portaudio ./configure make sudo make install make clean sudo apt-get install python3-pyaudio pip install pyaudio
-
Mac
sudo brew install libasound2-dev wget https://files.portaudio.com/archives/pa_stable_v190700_20210406.tgz tar -xvf pa_stable_v190700_20210406.tgz cd portaudio ./configure make sudo make install make clean pip install pyaudio
-
Windows
pip install pipwin pipwin install pyaudio
2.5 Install ImageMagick
You need to run video translation before installing
the official website download address
3. Settings
Modify config.conf
file:
- Set shortcut keys
- Set up the running device
gpu
orcpu
4. Run
Note: The first time the voice translation module needs to be connected to the Internet
python main.py
Show results
There are four functions: "Word Translation", "Screenshot Translation", "PDF Translation", and "Document Picture Translation". After the project is run, it will be mounted on the status bar. When you click "x", it will not exit but will be hidden. Click the status Click "Open" on the status bar and it will pop up. Click "Exit" on the status bar to actually exit the program. (Ubuntu18.04 and later have no status bar by default)
text translation
word translation
- Select "Chinese to English" or "English to Chinese" and click Start
- Then select the content you want to translate with your mouse and click copy
- Press the set shortcut key to translate
screenshot translation
Voice translation
Support audio files and voice recording
video translation
Support output srt subtitle files and videos
Document translation
- TXT
- Excel
- Word
Document image translation