Building:

cd data/image2text_docker
./build.sh

Starting:

cd data/image2text_docker
./start.sh path/to/image-to-text

Testing:

cd data/image2text_docker
./test.sh

For the whole speech-text-image-text-speech pipeline you can move data/image2text_docker/narration.sh and data/image2text_docker/text2speech to the top of image-to-text, download corresponding repositories and launch it.

https://github.com/tiiuae/image-to-text
https://github.com/tiiuae/voicerecognition/
https://github.com/tiiuae/CameraFetcher