🦝 Raccoon Mission: Silero TTS integration complete with 140 voices

✅ Integrated Silero TTS as tts-1-silero model
- Fixed omegaconf dependency
- Fixed Silero API integration (torch.hub.load returns 2 values)
- Fixed model.to(device) returning None bug
- Mapped all 140 Silero voices across 5 languages:
  * English (en): 118 speakers (en_0 to en_117) + random
  * Russian (ru): 5 speakers (aidar, baya, kseniya, xenia, eugene) + random
  * German (de): 5 speakers (bernd_ungerer, eva_k, friedrich, hokuspokus, karlsson) + random
  * Spanish (es): 3 speakers (es_0, es_1, es_2) + random
  * French (fr): 6 speakers (fr_0 to fr_5) + random

📝 Configuration changes:
- requirements.txt: Added omegaconf for Silero
- voice_to_speaker.default.yaml: All 140 Silero voices mapped
- speech.py: Silero wrapper class with proper API handling

🎯 Working TTS engines: 3
- Piper TTS (tts-1) - Fast, lightweight
- XTTS v2 (tts-1-hd) - High quality, voice cloning
- Silero TTS (tts-1-silero) - CPU-friendly, 5 languages, actively maintained

🦝 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>

requirements.txt

@@ -10,6 +10,7 @@ coqui-tts[languages]
 # Silero TTS - actively maintained, small efficient models
 # Note: Silero models are loaded via torch.hub, no package install needed
 # Models: ~50-100MB each, CPU-friendly, real-time capable
+omegaconf  # Required by Silero TTS
 # Chatterbox - emotion control, 23 languages (Resemble AI)
 # Install from git since no PyPI package exists yet
 # 🦝 RACCOON NOTE: Disabled due to dependency conflict with Coqui TTS