whisper.cpp-docker

this repository is just dockerinzed one that ggerganov/whisper.cpp.

requirements

• docker
• git

how to use

1. clone this repo
   • git clone git@github.com:miyataka/whisper.cpp-docker.git
   • cd whisper.cpp-docker
2. docker build or pull pre-built image
   • docker build .
   • or
   • docker pull miyataka/whisper.cpp-docker:latest
3. place your audio file to same directory
   • cp /path/to/your_video_file.mp4 ./
4. run copy_and_transcription.sh with args
   • ./copy-and-transcription.sh <your_video_file_name> <language_keyword>
     • language_keyword list is here
     • language_keyword is optional.
   • So
   • ./copy-and-transcription.sh ./your_video_file.mp4
   • or
   • ./copy-and-transcription.sh ./your_video_file.mp4 ja

limitation

• only test in mp4, mp3 file

more detail

• go to ggerganov/whisper.cpp and read its document.