Add Support for Audio Generation Orchestrator

[nickna]

Description

Extend the newly refactored MediaGenerationOrchestrator base class to support audio generation workflows.

Background

The recent refactoring created a generic MediaGenerationOrchestrator<TRequest, TResponse, TEventRequest> base class that successfully handles both image and video generation. This same pattern can be extended to support audio generation.

Requirements

1. Create AudioGenerationOrchestrator

• Extend MediaGenerationOrchestrator base class
• Implement abstract methods for audio-specific logic
• Handle audio-specific formats (MP3, WAV, OGG, etc.)
• Support both TTS and music generation workflows

2. Audio-Specific Processing

• Implement audio file validation
• Support streaming audio generation
• Handle audio metadata (duration, bitrate, sample rate)
• Implement audio transcoding if needed

3. Event Support

• AudioGenerationRequested
• AudioGenerationStarted
• AudioGenerationProgress
• AudioGenerationCompleted
• AudioGenerationFailed
• AudioGenerationCancelled

4. Provider Integration

• Support ElevenLabs provider (already in ProviderType enum)
• Support Ultravox provider (already in ProviderType enum)
• Consider adding OpenAI TTS support

Implementation Notes

• Follow the same patterns used in ImageGenerationOrchestrator and VideoGenerationOrchestrator
• Reuse existing media storage infrastructure
• Consider audio-specific cost calculation (per character/second)
• Add appropriate unit tests following MediaGenerationOrchestratorTestBase pattern

Benefits

• Unified architecture for all media generation
• Consistent error handling and retry logic
• Reusable infrastructure and patterns
• Easy to maintain and extend