Commit History
-
fix: Resolve syntax error in WebSocket retry logic
Peter Michael Gits +1: Claude- Fixed orphaned except blocks in _connect_websocket_with_retry method - Removed duplicate exception handling left from refactoring - Resolves SyntaxError at line 573 causing app startup failure - Update to v0.8.5 with critical syntax fix The error occurred when retry logic was moved to separate method but old exception handlers weren't properly cleaned up. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>34964d9
-
feat: Add comprehensive WebSocket retry logic with GPU cold start timing calibration
Peter Michael Gits +1: Claude- Add sophisticated exponential backoff retry mechanism (8 attempts, 3-45s delays) - Implement detailed timing calibration for GPU cold start pattern analysis - Add comprehensive status code analysis (503=cold start, 403=unavailable, etc.) - Track connection timing thresholds: <8s=warm, <30s=normal, >60s=loading issues - Add retry attempt logging with elapsed time tracking for performance debugging - Provide calibration summaries to identify service state patterns - Update to v0.8.4 with enhanced WebSocket resilience This addresses the cold start timing issues where STT services need time to warm up and provides detailed metrics to calibrate optimal retry parameters. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>ee023ba -
feat: Add STT service preload to prevent cold start delays
Peter Michael Gits +1: Claude- Add _preload_stt_service() method to warm up STT model during VoiceCal startup - Call /preload?model_size=base endpoint to trigger model loading - Prevent WebSocket HTTP 503 errors caused by cold-starting STT service - Add timeout handling and detailed logging for preload process - Update to v0.8.3 with STT service preload warming Resolves the timing issue where VoiceCal tries to connect before STT service is ready. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>628e022 -
feat: Update to v0.8.2 with WebSocket STT integration
Peter Michael Gits +1: Claude- Updated version to reflect WebSocket integration improvements - Enhanced description to include WebSocket STT and fallback mechanisms - Tracks the fix for WebSocket 403 Forbidden errors on HF Spaces 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>4a7e625 -
fix: Resolve Gradio client FileData validation error
Peter Michael Gits +1: Claude- Use handle_file() to create proper FileData objects for Gradio API calls - Fixes "meta field must be explicitly provided" validation error - Ensures Gradio client passes correctly formatted file metadata 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>f677de5 -
feat: Update version to v0.8.0 - Whisper corruption detection support
Peter Michael Gits +1: Claude- Updated version from 0.7.2 to 0.8.0 - Enhanced description to reflect STT service improvements - VoiceCal now supports STT service with Whisper corruption detection - Compatible with new factory reset and alternative STT models 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>ce0e860
-
chore: Bump version to v0.7.1
Peter Michael Gits +1: Claude- Update version for MCP fallback improvements - Build timestamp: 2025-08-22T15:00:00 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>e9456c7 -
feat: Implement unmute.sh streaming STT methodology v0.7.0
Peter Michael Gits +1: ClaudeUNMUTE.SH METHODOLOGY IMPLEMENTATION: - Buffer audio chunks during recording (like unmute.sh streaming) - Apply flush trick when recording stops - combine all chunks into complete audio - Process complete audio with is_final=True for proper transcription finalization - Clear buffer after processing to prevent chunk contamination - Add flush trick markers and metadata following unmute.sh patterns Key Changes: ✅ Buffer chunks during recording instead of processing individually ✅ Implement flush trick: combine chunks + process with is_final=True ✅ Send only final transcription after complete audio processing ✅ Clear buffers after each recording session ✅ Add unmute.sh-style metadata (is_final, chunks_processed) This should fix the 'Thank you very much' issue by properly implementing unmute.sh's streaming methodology with flush trick for clean transcriptions. 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>7d32abe -
debug: Add audio debugging to trace 'Thank you very much' source v0.6.1
Peter Michael Gits +1: ClaudeDEBUGGING: Add detailed logging to find source of repeated transcription - Add audio file size and hex dump debugging in WebSocket STT calls - Add audio tuple content debugging in Gradio audio processing - Log exact audio data being sent to STT service - This will help identify if voiceCal is sending wrong/corrupted audio - or if STT service has internal issues with specific audio content 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>b61f866 -
feat: Remove all demo simulation code - REAL STT ONLY v0.6.0
Peter Michael Gits +1: ClaudeMAJOR CHANGE: Complete removal of demo/simulation modes - Remove all _simulate_stt() and _simulate_stt_with_length() methods - Remove demo_mode flags and logic throughout MCPAudioHandler - Force voiceCal to only use real STT services (WebSocket/MCP/HTTP) - Return clear error messages when STT services fail instead of fake responses - Update version to 0.6.0 for this breaking change - No more confusing demo transcriptions that mask real issues Real Services Only: ✅ WebSocket STT (primary) ✅ MCP STT (fallback) ✅ HTTP STT (final fallback) ❌ No demo/simulation modes 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>4cf4588 -
fix: Remove max_file_size parameter for HF Spaces compatibility v0.5.11
Peter Michael Gits +1: ClaudeCRITICAL FIX: Resolve voiceCal deployment crash - Remove max_file_size parameter from gr.Blocks() - not supported in HF Spaces Gradio version - Fix TypeError: BlockContext.__init__() unexpected keyword argument error - Update to v0.5.11 to trigger fresh deployment - Add compatibility note for future reference - This should resolve the exit code: 1 deployment failure 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>9a4b253 -
fix: Improve WebSocket stability and prevent spinning v0.5.10
Peter Michael Gits +1: Claude- Update version to v0.5.10 to trigger fresh deployment - Add better timeout handling for WebSocket connections (15s connect, 30s response) - Disable ping_interval to reduce connection overhead - Improve timeout handling to prevent service from hanging or spinning - Force deployment refresh to resolve any stuck container issues 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>54cc854
-
fix: Add max_file_size to resolve Gradio upload error v0.5.9
Peter Michael Gits +1: Claude- Fix AttributeError: 'Blocks' object has no attribute 'max_file_size' - Set 10MB max file size for audio uploads - Addresses 500 errors in Gradio upload endpoint 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>edd02a0 -
fix: Add Hugging Face Spaces GPU support v0.5.8
Peter Michael Gits +1: Claude- Add required 'import spaces' and @spaces.GPU decorator - Create initialize_gpu_resources() function to satisfy HF Spaces requirement - Add spaces>=0.19.0 to requirements.txt - Update to version 0.5.8 Fixes "Runtime error no @spaces.GPU function detected during startup" 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>b63ba79 -
fix: Add PyTorch dependencies and fix base64 imports v0.5.7
Peter Michael Gits +1: Claude- Add missing PyTorch, TorchAudio, and Transformers dependencies to requirements.txt - Fix missing base64 import statements in WebSocket handler - Add better HTTP 503 error handling for cold starts - Update to version 0.5.7 with comprehensive fixes 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>e985487
-
fix: Resolve Gradio chatbot warning and 404 errors v0.5.6
Peter Michael Gits +1: Claude- Fix deprecated chatbot type parameter (set type='messages') - Update version display to v0.5.6 using dynamic import - Mount Gradio interface on FastAPI to fix 404 errors - Hybrid deployment: WebRTC endpoints + full Gradio interface - Clean up duplicate code and improve error messages 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>cb07c5a -
feat: Deploy complete VoiceCal application with all files v0.5.6
Peter Michael Gits +1: Claude- Add all application files: app.py, requirements.txt, core modules - Include WebRTC integration and voice service components - Complete Gradio application ready for Hugging Face deployment - Comprehensive README with VoiceCal branding and documentation - All dependencies and Docker configuration included 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>5e8a657