GenAIExamples/Translation

Language Translation

Language Translation is the communication of the meaning of a source-language text by means of an equivalent target-language text.

The workflow falls into the following architecture:

Start Backend Service

1. Start the TGI service to deploy your LLM

cd serving/tgi_gaudi
bash build_docker.sh
bash launch_tgi_service.sh

launch_tgi_service.sh by default uses 8080 as the TGI service's port. Please replace it if there are any port conflicts.

2. Start the Language Translation service

cd langchain/docker
bash build_docker.sh
docker run -it --name translation_server --net=host --ipc=host -e TGI_ENDPOINT=${TGI_ENDPOINT} -e HUGGINGFACEHUB_API_TOKEN=${HUGGINGFACEHUB_API_TOKEN} -e SERVER_PORT=8000 -e http_proxy=${http_proxy} -e https_proxy=${https_proxy} translation:latest bash

Here is the explanation of some of the above parameters:

• TGI_ENDPOINT: The endpoint of your TGI service, usually equal to <ip of your machine>:<port of your TGI service>.
• HUGGINGFACEHUB_API_TOKEN: Your HuggingFace hub API token, usually generated here.
• SERVER_PORT: The port of the Translation service on the host.

3. Quick test

curl http://localhost:8000/v1/translation \
    -X POST \
    -d '{"language_from": "Chinese","language_to": "English","source_language": "我爱机器翻译。"}' \
    -H 'Content-Type: application/json'

The shortcodes of languages are also supported:

curl http://localhost:8000/v1/translation \
    -X POST \
    -d '{"language_from": "de","language_to": "en","source_language": "Maschinelles Lernen"}' \
    -H 'Content-Type: application/json'