ElevenLabs Announces "Conversational AI" — From Voice Synthesis to Dialogue Agents
ElevenLabs announced “Conversational AI,” a developer platform for building conversational voice agents. Rather than simply converting text to speech, it now enables developers to build complete voice agents from the ground up — a turning point that significantly broadened the company’s scope.
Details
- What it enables: Developers can build a complete conversational agent on the platform, customizing parameters like tone and response length
- A shift in positioning: This marked the first flagship product showing the company’s evolution from a voice synthesis API provider into a company handling entire conversations as an AI agent
- Intended use cases: Aimed at use cases like customer support bots and voice assistants, where companies want to embed their own voice agents
- Development foundation: Built on the technical foundation the company developed in voice synthesis, multilingual support, and dubbing, now newly integrated with dialogue management capabilities
What happened next
The launch of Conversational AI became the starting point for ElevenLabs’ entry into the agent market, moving beyond voice synthesis. In February 2025, the company introduced its speech recognition model “Scribe,” rounding out both the “generation” and “recognition” sides of voice as it moved toward becoming a comprehensive voice AI company.