Streaming TTS Explained

HostFabric uses two different audio playback systems depending on your TTS provider. Streaming TTS (ElevenLabs) starts playing immediately, while traditional TTS waits for complete audio generation.

• Check your TTS provider in Settings → Voices
• Look for "streaming" indicator in debug mode

With ElevenLabs, responses start playing within 300-600ms. This creates natural conversation flow where AI can interrupt or respond quickly to live chat.

• Switch to ElevenLabs TTS if not already using it
• Send a short prompt and notice the fast response
• Try interrupting with another prompt mid-response

Google TTS and OpenAI TTS use traditional methods with 2-4 second delays. Perfect for pre-recorded content but slower for live interaction.

• Switch to Google TTS to see the difference
• Notice the longer wait before audio starts
• See how responses feel more "robotic" in timing

Choose ElevenLabs for live, interactive shows. Use Google/OpenAI TTS for tutorials, presentations, or when perfect lip-sync is critical.

• Pick TTS based on your content type
• Test both methods during rehearsal
• Consider having backup TTS configured