How tauhid-topu-007/verba-translator works
Verba Translator lets a user enter text, choose a target from 25 languages, and request a translation. The Streamlit client validates nonempty input, posts it to the FastAPI invoke endpoint, and displays the returned translation using session state. The backend validates the request, runs a LangChain prompt/model/parser chain, and returns text generated through Groq. A separate health endpoint is available to API callers. The graph reflects the README and sampled source; no persistent translation history or other future features are shown.
Main entry points: serve.py FastAPI Routes · client.py Request Client
Components 12
Translation UI
- Streamlit Interface
client.py - Request Client
client.py - Translation Session
client.py
Translation Backend
- FastAPI Routes
serve.py - Request Validation
serve.py - Translation Chain
serve.py - Language Prompt
serve.py - ChatGroq Model
serve.py - Text Parser
serve.py
Model Provider
- Groq API
Other components
- User
- API Caller
Connections 15
- User → Streamlit Interface: enters text
- Streamlit Interface → Request Client: submits request
- Request Client → FastAPI Routes: POST invoke
- FastAPI Routes → Request Validation: validates with
- FastAPI Routes → Translation Chain: invokes
- Translation Chain → Language Prompt: formats input
- Language Prompt → ChatGroq Model: provides messages
- ChatGroq Model → Groq API: requests completion
- ChatGroq Model → Text Parser: returns response
- Text Parser → Translation Chain: returns text
- Translation Chain → FastAPI Routes: returns translation
- FastAPI Routes → Request Client: returns output
- Request Client → Translation Session: stores translation
- Translation Session → Streamlit Interface: renders result
- API Caller → FastAPI Routes: GET health