Handler for the /speak endpoint

Example request

{
  "text": "Hello, world!",
  "tts_config": {
    "audio_format": "linear16",
    "connection_timeout": 30,
    "model": "aura-asteria-en",
    "pronunciations": [
      {
        "pronunciation": "A P I",
        "word": "API"
      }
    ],
    "provider": "deepgram",
    "request_timeout": 60,
    "sample_rate": 24000,
    "speaking_rate": 1,
    "voice_id": "aura-asteria-en"
  }
}