Add FluidAudio to Speech Processing section

[Alex-Wengg]

Summary

Adds FluidAudio to the Speech Processing section.

FluidAudio is a Swift SDK that brings frontier audio AI models to Apple devices through CoreML integration. It provides:

• Automatic Speech Recognition (ASR) - Using NVIDIA's Parakeet TDT model, supporting 25 European languages
• Speaker Diarization - Streaming and offline modes for identifying multiple speakers
• Voice Activity Detection (VAD) - Using Silero models
• Text-to-Speech (TTS) - Using the Kokoro model

All models run fully on-device using the Apple Neural Engine (ANE) for low-latency, privacy-preserving audio AI.