How nnavyy/livetranslatoroverlay works
Live Translator Overlay is a Windows desktop app for live audio subtitles and selected-region OCR translation, with history, localization, and in-app report tracking. The main workflow starts in the overlay: capture audio or select a screen region, transcribe or recognize text, translate it, then display the result. README capabilities supplement the sampled source; links to unsampled subsystems are limited to documented behavior and supported paths.
Main entry points: MainWindow.xaml.cs Translator overlay · WhisperSttService.cs Batch transcription · FeedbackHistoryManager.cs Report history manager · GroqTranslateProvider.cs Groq translation
Components 24
Desktop experience
- Translator overlay
MainWindow.xaml.cs
Audio transcription
- Audio capture
Core/AudioCaptureService.cs - Streaming diarization
Core/DiarizedSttService.cs - Batch transcription
Core/WhisperSttService.cs - Transcript filtering
Core/HallucinationDetector.cs
Text processing
- Screen OCR
Core/ScreenOcrService.cs - Google translation
Core/Translation/GoogleFreeTranslateProvider.cs - Groq translation
Core/Translation/GroqTranslateProvider.cs - Libre translation
Core/Translation/LibreTranslateProvider.cs - Translation provider contract
Core/Translation/ITranslationProvider.cs
History and feedback
- Translation history
HistoryWindow.xaml.cs - Feedback interface
FeedbackWindow.xaml.cs - Report history manager
Core/Feedback/FeedbackHistoryManager.cs - Report record
Core/Feedback/FeedbackRecord.cs - Local report history
Local configuration
- App settings
Core/AppSettings.cs - UI localization
Core/Localization/LocalizationManager.cs - Quota tracking
Core/QuotaTracker.cs
Other components
- Desktop user
- Audio devices
- Desktop screen
- Deepgram / Groq APIs
- Translation services
- GitHub report status
Connections 31
- Desktop user → Translator overlay: controls
- Translator overlay → Audio capture: starts capture
- Audio devices → Audio capture: provides audio
- Translator overlay → Streaming diarization: streams audio
- Translator overlay → Batch transcription: submits audio
- Streaming diarization → Deepgram / Groq APIs: streams to
- Batch transcription → Deepgram / Groq APIs: transcribes via
- Streaming diarization → Transcript filtering: filters transcript
- Batch transcription → Transcript filtering: filters transcript
- Translator overlay → Screen OCR: recognizes region
- Desktop screen → Screen OCR: supplies region
- Translator overlay → Google translation: requests translation
- Translator overlay → Groq translation: requests translation
- Translator overlay → Libre translation: requests translation
- Google translation → Translation services: translates via
- Groq translation → Translation services: translates via
- Libre translation → Translation services: translates via
- Translation provider contract → Google translation: defines contract
- Translation provider contract → Groq translation: defines contract
- Translation provider contract → Libre translation: defines contract
- Translator overlay → Translation history: opens history
- Translator overlay → Feedback interface: opens reports
- Feedback interface → Report history manager: loads and updates
- Report history manager → Report record: creates records
- Report history manager → Local report history: reads and writes
- Report history manager → GitHub report status: checks status
- Translation history → Groq translation: summarizes history
- Translator overlay → App settings: loads preferences
- Translator overlay → UI localization: localizes interface
- Translator overlay → Quota tracking: updates usage
- Batch transcription → Quota tracking: updates quota