You can use the Audio Management tab to enhance the quality of your assistant’s voice by for commonly-used TTS (Text-to-Speech) audio elements, like greetings or transfer messages. This ensures consistent voice quality while reducing latency during interactions.

​
Benefits

  • Latency reduction: Serve cached audio to minimize TTS latency.
  • Improved audio quality: Dependable, reliable high-quality TTS generation using the cache.
  • Consistency: Maintain uniform voice quality and brand voice and tone across all your assistant interactions.

​
Getting started

To manage your assistant’s audio:

  1. Go to the Audio Management tab.
  2. Review all audio saved to the cache and monitor how often it has been used by the assistant.
  3. In future updates, you’ll be able to delete cached files and upload new ones to overwrite existing audio.

You can edit the stability and clarity of the assistant’s voice specifically for this utterance. For more information on these settings, visit the voice configuration feature page. The edit tab also includes sync and play buttons so you can test changes to the utterance live in the edit panel. .

​
Interaction style

The Interaction style settings feature lets you adjust your assistant’s response latency settings. Customize how quickly the assistant replies to balance speed and accuracy based on your project’s requirements.

​
Benefits

  • Customizable response time: Choose from predefined modes tailored to your needs.
  • Enhanced user experience: Minimize wait times or maximize response precision to optimize interactions.

​
Interaction style settings

  1. Locate the Interaction style section in the Settings menu.
  2. Choose from the available modes:
    • Swift Mode
    • Balanced Mode
    • Precise Mode
    • Turbo Mode
  3. Click on the bubble for your preferred mode. A brief description of the mode will appear.
  4. Save your settings to apply changes. Your assistant will adjust its behavior immediately.

​
Performance characteristics

Each response mode is designed for specific performance needs:

​
Turbo

  • Latency: 400ms
  • Description: This is our ultra-fast mode, designed to make the assistant more responsive by reducing the wait time before it responds to a caller. However, this increased speed may cause the assistant to interrupt more frequently. To address this, we recommend enabling barge-in when using turbo mode. Barge-in helps balance the assistant’s rapid responses by allowing callers to regain control seamlessly.

​
Swift

  • Latency: 1000ms
  • Description: Prioritizes speed, providing the fastest response times with higher interruption tolerance. Best for quick interactions where speed is more valuable than precision.

​
Balanced

  • Latency: 1400ms
  • Description: Offers moderate speed and balanced interruption levels. Ideal for general use cases requiring a compromise between responsiveness and accuracy.

​
Precise

  • Latency: 1800ms
  • Description: Delivers the most accurate responses with minimal interruptions, at the cost of slower response times. Suitable for scenarios where accuracy is paramount.

​
Barge-in

The barge-in settings toggle allows you to decide if callers can interrupt assistants. This can improve conversation flow and create natural, human-like interactions, but it can also cause disruption to the call experience, depending on your user needs. Experiment with barge-in and latency modes to find a conversational rhythm that suits you.

Functionally, this feature shortens the Voice Activation Detection (VAD) time and reduces response latency.

To test this feature, find “Enable barge-in” in the “Settings” menu.

ModelEnvironments & versions