Neural Voice Synthesis enables the platform to speak translated text aloud in natural, human-like voices. It transforms written or transcribed translations into clear, expressive speech — making conversations more inclusive for audiences who prefer listening over reading, or for situations where reading is impractical.

The system offers a range of voice profiles across different genders, styles, and accents, allowing you to choose a delivery that best suits your audience and context. For example, you might select a warm, friendly voice for customer service, or a formal, authoritative tone for official announcements.

Neural Voice Synthesis works across In-Person, Video Call, and Broadcast modes. Once enabled, the translated text is instantly converted into audio in the listener’s preferred language. For live events, the voice output is synchronised with captions to keep the spoken and written translations aligned.

Key benefits include:

  • Accessibility for low-literacy users or those with vision impairments
  • Engagement through natural intonation and pacing
  • Flexibility to match tone and style to the situation

In the following sections, you’ll learn how to:

  • Enable or disable neural voice playback
  • Select and customise voice profiles
  • Combine voice synthesis with glossary and prompt controls
  • Optimise clarity and timing for live speech translation