For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Start buildingGet support
DocumentationAPI ReferenceChangelogDiscord
  • Introduction
    • Welcome to Hume AI
    • Getting your API keys
    • Support
    • Pricing
  • Voice
    • Overview
    • Voice design
    • Voice cloning
    • Voice management
  • Text-to-Speech (TTS)
    • Overview
    • Voice
    • Acting instructions
    • Voice conversion
    • Continuation
    • Timestamps
    • FAQ
  • Speech-to-Speech (EVI)
    • Overview
      • Interruptibility
      • Pause responses
      • Dynamic variables
      • Context injection
      • Tool use
      • Resume chats
      • Chat history
      • Audio reconstruction
    • FAQ
  • Expression Measurement
    • Overview
    • About the science
    • FAQ
  • Integrations
    • MCP
    • Vercel AI SDK
    • LiveKit
    • Pipecat
    • Vapi
    • Twilio
    • Agora
  • Resources
    • Terms of use
    • Use case guidelines
    • Billing
    • Errors
    • Privacy
    • Status
Start buildingGet support
LogoLogo
LogoLogo
On this page
  • How interruption works
  • Handling interruptions client-side
  • Configuring interruption sensitivity
Speech-to-Speech (EVI)Features

Interruptibility

Guide to EVI's interruptibility feature and how to manage interruptions on the client.
Was this page helpful?
Edit this page
Previous

Pause Responses

Guide to pausing EVI's responses during a Chat session.
Next
Built with

Interruptibility is a core feature of EVI that enables natural, real-time interaction. Users can speak over the assistant at any time. When they do, EVI detects the interruption, stops generating the response, and immediately notifies the client.

How interruption works

EVI continuously processes incoming audio, even while generating responses. An interruption occurs when speech is detected while the assistant is responding. Upon detecting an interruption, EVI:

  • Stops generating the current response
  • Stops streaming response audio
  • Sends a user_interruption message to the client.

Handling interruptions client-side

To make the interruption perceptible, the client must stop audio playback.

Although EVI halts response generation, the user won’t experience the interruption unless the assistant’s voice also stops.

When a user_interruption message is received, the client should:

  1. Stop audio playback: Halt any ongoing playback of the interrupted response.
  2. Clear queued audio: Discard any queued audio from the previous assistant response.

Interruption is handled automatically when using the React SDK.

Configuring interruption sensitivity

By default, EVI interrupts as soon as it detects user speech. To adjust how much speech is required before EVI yields, see the Interruption configuration guide.