Hume AI

Hume AI is a research laboratory dedicated to the development of artificial intelligence systems with empathetic capabilities, aiming to create multimodal AI technologies that possess emotional intelligence. They offer advanced AI models such as Octave Text-to-Speech (TTS), which is the first large-scale language model capable of understanding context and predicting emotions in text-to-speech conversions; as well as Empathic Voice Interface (EVI), a real-time, customizable voice intelligence system that enables smooth and emotionally expressive conversations. Additionally, they provide an expression analysis API that allows for the detection of emotional cues in facial expressions, voices, and language. Their goal is to create AI systems that deliver expressive voices and interactive personalities, with a particular focus on promoting human well-being and the ethical development of artificial intelligence technologies.

Users can utilize Octave TTS to generate AI-generated voices by providing textual prompts and specifying the desired voice characteristics, attributes, and emotional tone. They can also use EVI to create and interact with real-time synthesized voices and personalized audio profiles, enabling flexible adjustments to prompts and pitch settings. Developers have access to APIs and a comprehensive developer platform, allowing them to integrate these emotionally intelligent voice agents into their own applications.

Octave Text-to-Speech (TTS): A large language model that is capable of understanding context, predicting emotions, and allowing the control of emotional expression and speaking style through natural language commands.
Octave Voice Design: It is possible to create any AI-generated voice with just a few simple instructions or a vivid script.
Empathic Voice Interface (EVI): A real-time, customizable voice-based intelligent system designed for a wide range of applications. It enables smooth conversations, understands the user’s tone, and generates responses in a manner that matches that tone appropriately.
Expression Measurement API: Used to analyze expressions detected in facial features, vocalizations, and speech.
Conversational Voice: A comprehensive developer platform for deploying voice agents with emotional intelligence.
Sound modulation: Adjusts the basic sound of EVI 2 across a continuous range of settings, including feminizing the tone, adding a nasal accent, or changing the pitch.
EVI’s emerging multilingual capabilities.