Skip to content

Introducing Expressive Mode for ElevenAgents

More expressive voice agents, built for real-world customer conversations.

expressive mode

Today we’re introducing Expressive Mode for ElevenAgents - voice agents so expressive, they blur the line between AI and human conversations. 

Expressive Mode gives teams unprecedented control over tone - so agents can de-escalate, reassure, and guide conversations to a clear resolution, even under pressure.

This is an unedited recording of an agent empathizing with a customer at peak frustration after a flight cancellation:

mark screenshot w caption space

Two core upgrades powering Expressive Mode

It is powered by two upgrades:

  • Eleven v3 Conversational - our most emotionally intelligent, context-aware Text to Speech model, built on Eleven v3 and optimized for real-time dialogue.
  • A new turn-taking system - better-timed responses with fewer interruptions.

These releases were developed in parallel to fit seamlessly together within ElevenAgents  -  our vertically integrated conversational agents platform.

Eleven v3 Conversational

Eleven v3 Conversational is our most emotionally intelligent, context-aware TTS model to date  -  maintaining conversational context across turns and delivering speech that reflects intent, emotion, and emphasis without sounding scripted or exaggerated.

For example, you can guide an agent to use a calmer, more reassuring tone when a user sounds worried  -  or a more direct tone when clarity and speed matter. These controls help teams align delivery with brand voice and user expectations.

A new turn-taking system

Human conversation depends as much on timing as on words. Interrupt too early and you break trust. Respond too late and the moment is gone.

Our new turn-taking system uses real-time signals from our industry-leading transcription model, Scribe v2 Realtime, to infer emotion from how something is said and better determine when an agent should speak, pause, or wait. This helps agents respond more like humans do, especially in emotionally charged situations.

A sudden rise in speaking pace - accompanied by volume surges and repetition - often signals acute stress or panic:

By contrast, bright rising intonation with short exclamations often aligns with pleasant surprise or relief:

Built for global customer operations

Expressive Mode scales emotional nuance across 70+ languages, improving delivery in languages and dialects where nuance previously lagged, such as Hindi.

Hear an agent speaking various Indian languages:

 / 

And Spanish with an Argentinian accent:

 / 

Try Expressive Mode today

Deploy emotionally expressive agents that stay on-brand, respond at the right moment, and adapt to real customer emotion - so customers feel helped, not processed.

Built for production in ElevenAgents, with the reliability, integrations, testing, and monitoring required for large-scale customer operations - from pilot deployments to global rollouts.

Try it out: Live demo

Read the docs: Docs

Explore articles by the ElevenLabs team

ElevenLabs

Create with the highest quality AI Audio

Get started free

Already have an account? Log in