LocalAI & easy Piper compatibility release (0.3.5)

Now supporting LocalAI (https://localai.io/)

• Improved Audio Format Detection: Enhanced TTS audio streaming with automatic WAV header parsing to detect correct sample rate, channels, and bit depth for LocalAI
• Docker Compose Templates: Added docker-compose.localai.yml for easy LocalAI deployment
• LocalAI supports both Piper and Whisper. An example compose is provided with the following defaults:
  • piper:
    • en-us-kathleen-low
    • en-us-amy-low
    • en-us-danny-low
    • en-us-lessac-low/medium
    • en-us-libritts-high
    • en-us-ryan-high/low/medium
  • whisper-base

Full Changelog: v0.3.4...v0.3.5