This script chains together a bunch of AI/ML models to create audio clips separate by sentences that can be used as Anki cards The output from the script is audio clips, and a txt file that can be used by Anki to import the cards The front of the card is the audio, the back is the transcription and translation. All of the AI tools run locally. No payments required
use the requirements.txt
ex: pip install -r requirements.txt
Make sure you have ffmpeg installed. It's used for splitting the videos.
- For Spacy languages or GPU usage check https://spacy.io/usage
- Argo Translate https://github.com/argosopentech/argos-translate/tree/master
- Whsiper https://github.com/openai/whisper/tree/main
- Anki File locations https://docs.ankiweb.net/files.html#file-locations
Generate clips from a video, save the audio clips and generate a txt file that can be imported into Anki
eg. ./video_2_anki --audio-save-dir=/path/to/Anki2/User\ 1/collection.media foreign_language_video.mp4
Or download straight from YouTube
eg. ./video_2_anki --audio-save-dir=/path/to/Anki2/User\ 1/collection.media https://www.youtube.com/watch?v=some-video-id
This will save all the videos to the path, using the Anki path means you can just import the txt file.
Txt file example:
[sound:slow_spanish-0-clip.mp4];Hola a todos.<br>Hello, everyone.
[sound:slow_spanish-1-clip.mp4];¿Cómo están?<br>How are you?
./video_2_anki --help for more detailed options
Only supports Spanish to English, but other languages can be added!