Commit Graph
4 Commits
Author SHA1 Message Date
diodiogod 02cbc451fa Clean up root directory - move RVC docs and remove test file 2025-08-08 14:56:08 -03:00
diodiogod 6ae4ad15ef Improve user experience with clear tooltips and remove misleading widgets
• Enhanced tooltips across all RVC nodes with technical terms + user explanations
• Removed misleading output_format widget from Merge Audio (format handled by Save Audio)
• Added quality-loss warnings for audio processing operations (stereo conversion, resampling, padding)
• Updated PROJECT_INDEX.md to document complete RVC implementation
• Moved reference project to IgnoredForGitHubDocs (works independently now)
• Added lesson to CLAUDE.md about keeping dependencies when copying implementations

Technical improvements:
- Clear tooltips: "Consonant protection - Low=voice changes more, High=clearer pronunciation"
- Pitch controls: "deeper voice like man" vs "higher voice like child" (non-technical language)
- Smart warnings only for significant quality loss (>0.1s padding/truncation, stereo conversion)
- Torch tensor processing remains lossless between ComfyUI nodes
2025-08-08 08:32:26 -03:00
diodiogod 34c7a823c2 Version 4.1.0: Add RVC integration with iterative refinement and caching 2025-08-08 00:34:34 -03:00
diodiogodandSayanoAI cf25269595 Add RVC Engine integration with comprehensive voice conversion support
Implement RVC (Real-time Voice Conversion) as unified engine in TTS Audio Suite:

• RVC Engine Node: Consolidates RVC model, Hubert model, and voice conversion parameters into single interface
• RVC Pitch Extraction Options: Advanced pitch settings for professional control
• Vocal Removal Node: UVR5-based audio source separation for vocals/instrumentals
• Merge Audio Node: Advanced audio mixing with multiple algorithms
• Voice Changer Integration: Extended to support both ChatterBox and RVC engines

Architecture Features:
• Modular engine adapter following existing TTS Suite patterns
• Intelligent model management with caching and auto-discovery
• Seamless audio format conversion between ComfyUI and RVC formats
• Progressive complexity (simple presets to advanced controls)
• Professional error handling with graceful degradation

Integration Benefits:
• Compatible with Character Voices and existing TTS workflow
• Supports complex audio production workflows
• Foundation for future RVC enhancements and model training

Co-authored-by: SayanoAI <noreply@github.com>
2025-08-07 21:18:30 -03:00