Cleo brings voice to its AI financial assistant with ElevenLabs

Cleo is an AI-powered financial assistant used by millions of people to manage their money through chat. Known for its distinctive personality and direct tone, Cleo helps users budget, save, build credit, and access cash advances.

Cleo's chat-based interface already made financial management feel conversational, but typing still created friction for users, especially when they needed quick guidance on the go. To create a more seamless experience, Cleo used ElevenAPI to power its Voice Mode that lets users speak naturally with their financial assistant rather than typing.

Preserving Cleo’s personality through realistic, expressive voice

Cleo is known for being direct, witty, and occasionally blunt, with features like "Roast Mode" that calls out overspending. Translating that personality into spoken audio required a Text to Speech model expressive enough to carry the necessary tone, humor, and encouragement. With ElevenLabs, Cleo’s assistant delivers spoken responses that reflect the assistant's character rather than sounding flat or generic.

Personality traits that work well in text, such as sarcasm, encouragement, and directness, land with greater impact in voice, creating a stronger emotional connection with users and making financial nudges more effective.

Low-latency streaming across a global user base

Cleo serves millions of users worldwide. Adding voice at that scale requires both speed and consistency. ElevenAPI streams spoken responses fast enough to maintain natural conversational flow, while providing the reliability needed to deliver a consistent voice experience across Cleo's entire user base.

Voice as a growth lever

Cleo's Voice Mode shows how expressive, low-latency Text to Speech can add a new dimension to products that already rely on conversation. 

For developers and product teams building in fintech or conversational AI, ElevenAPI provides the tools to bring voice experiences like this to production at scale.