Emotive voice AI startup Hume launches new EVI 3 model with rapid custom voice creation
1 min read
Summary
NYC-based AI startup Hume has released its latest conversational AI model, dubbed Empathic Voice Interface (EVI) 3, which is designed to allow users to create their own voices simply by talking to the model, with the AI adjusting its own responses based on the user’s emotions.
The EVI 3 is aimed at developers and businesses, and offers more sophisticated customisation, swifter responses and more advanced emotional understanding than previous models, with internal tests showing that users preferred EVI 3 over competitors OpenAI’s GPT-4o and Google’s Gemini models.
While Hume’s models lack the voice cloning function offered by rivals, the company has said it will add the capability to its Octave text-to-speech model.
Access to EVI 3 for developers is set to be made available in the coming weeks via Hume’s API, with the company’s flexible pricing plan presumably set to be revealed at the same time.