How lokesh701449/voice-ai-assistant works
This is a browser-based voice assistant for recording or uploading audio, transcribing it, translating the transcript, and generating playable/downloadable speech. The principal flow runs from the browser UI through Flask endpoints to Whisper transcription or translation, then optionally to speech generation. The README contains screenshots but no textual capability description; the workflow is evidenced by the sampled source. The standalone speech script is included, but its integration with the web app is not shown.
Main entry points: app.py Flask routes · app.js Voice assistant UI · t.py Audio transcription · translate.py Text translation
Components 9
Browser experience
- Voice assistant UI
app.js
Web API
- Flask routes
app.py
Language processing
- Audio transcription
t.py - Text translation
translate.py
Speech generation
- Standalone speech tool
speech.py
Other components
- User
- Whisper library
- Google Translate
- gTTS service
Connections 11
- User → Voice assistant UI: records or uploads
app.js - Voice assistant UI → Flask routes: requests transcription
app.js - Voice assistant UI → Flask routes: requests translation
app.js - Voice assistant UI → Flask routes: requests speech
app.js - Flask routes → Audio transcription: transcribes audio
app.py - Audio transcription → Whisper library: uses model
t.py - Flask routes → Text translation: translates text
app.py - Text translation → Google Translate: requests translation
translate.py - Flask routes → gTTS service: generates speech
app.py - Voice assistant UI → Flask routes: requests audio download
app.js - Standalone speech tool → gTTS service: generates speech
speech.py