October 2025 Summaries

AudioStack is an AI-driven audio production platform utilized by major global creative teams to streamline and enhance the production of audio content such as advertisements and podcasts. By integrating Hume's emotionally intelligent voices into their voice library, AudioStack addresses the need for consistent stability and natural expressiveness in enterprise content generation. This integration allows for the production of audio that is both reliable and emotionally engaging, making it suitable for large-scale operations while significantly reducing costs. The platform's capability to scale across different markets and languages enhances its appeal to agencies, publishers, and brands aiming for high-quality, authentic audio content.

Journee, known as "The Human AI Company," specializes in creating natural AI interactions for global brands through customizable AI agents for sales, customer support, and engagement. Facing challenges with their previous voice provider, such as high costs, complex integrations, and scalability issues, Journee integrated Hume's Empathic Voice Interface (EVI). This transition reduced costs, simplified integration with clean APIs, and improved performance with low latency and expressive, natural voices, enabling Journee to deploy voice agents more efficiently and effectively across customer touchpoints. Hume's technology, which captures the nuances of human conversation, aligns with Journee's mission to create emotionally intelligent and accessible AI, transforming digital interactions into empathetic dialogues.

Render Foundry, led by Shane Boyce and Josh Harwell, specializes in creating immersive experiences using Unreal Engine, notably achieving a groundbreaking interactive experience featuring Babe Ruth with the aid of Hume's custom voice cloning technology. This project enables visitors to engage in authentic conversations with a virtual Babe Ruth, capturing his tonal qualities, cadence, and personality, thus transcending typical museum exhibits and offering a unique blend of education and personal connection. Through a seamless combination of visual and audio simulation, Render Foundry's work exemplifies the potential of empathic AI in creating richly interactive digital experiences that humanize AI responses, as noted by Creative Director Josh Harwell.

Niantic Spatial, in collaboration with Snap Inc. and Hume AI, is advancing the field of augmented reality by integrating spatial intelligence with emotionally aware AI to create location-aware companions for AR glasses. Their flagship project, Project Jade, features a companion named Dot, which guides users through the physical world using a deep understanding of context and natural dialogue. The initiative is built on three key pillars: first-person awareness through AR glasses, context-rich storytelling, and emotional intelligence, allowing Dot to deliver personalized and engaging interactions. Hume AI's Empathic Voice Interface (EVI) provides the emotionally adaptive voice that enables Dot to respond with enthusiasm or calmness depending on the user's engagement level. This partnership transforms digital maps from static tools into lively companions, enhancing user interaction with their environment by combining Niantic Spatial's geospatial AI, Snap's AR technology, and Hume's voice capabilities.

Revelum is an AI-native security platform founded by Enrique Barco in 2025, designed to protect institutions from deepfake impersonations and fraud in real time, using developer-friendly APIs and turn-key solutions. Partnering with Hume AI, Revelum leverages advanced detection systems capable of identifying sophisticated synthetic voices, including Hume's Empathic Voice Interface (EVI). A recent demo illustrated Revelum's technology effectively flagging a deepfake during an account takeover attempt by an AI voice agent, showcasing its real-time risk assessment and detection capabilities. The collaboration with Hume AI creates a feedback loop for responsible AI development, allowing Revelum to train and refine its models using Hume's advanced voice technology, thus staying ahead of emerging threats. This partnership emphasizes the proactive approach of anticipating the evolving tactics of malicious actors to ensure the responsible use of generative AI.

Inception Point, led by CEO Jeanine Wright, is revolutionizing storytelling by creating AI personalities that engage audiences across various media, including podcasts, video, and books. The company operates the Quiet Please Podcast Network, the world's largest independent podcast network by show count, producing over 3,000 episodes weekly for more than 11 million listeners. They manage over 50 AI personalities, each uniquely crafted to captivate listeners, and produce episodes for $1 or less, making podcasts profitable after just 20 listeners. Inception Point's expansion strategy involves integrating Hume's Empathic Voice Interface (EVI) to ensure their AI hosts can create emotionally intelligent and engaging content while maintaining high audio quality across diverse formats. This technology allows Inception Point to scale its operations without compromising on the expressiveness or quality of its AI-generated content, as highlighted by Chief Content Officer Katie Brown, who credits Hume with providing unmatched voice character and emotional depth.

Octave 2, the latest iteration of the voice AI model, enhances text-to-speech capabilities by understanding emotional tones and extending support to 11 languages, including Japanese and Korean. It generates audio swiftly at under 200ms and is offered at half the price of its predecessor, Octave 1. New features include voice conversion and direct phoneme editing, which allow for realistic voice swapping and precise pronunciation adjustments, respectively. Octave 2's efficiency is bolstered by advanced chip deployment and a novel inference stack, enabling large-scale application across industries like entertainment and customer service. The recently launched EVI 4 mini extends these capabilities to a speech-to-speech API, facilitating the creation of interactive experiences, though it currently requires pairing with an external language model for native language generation. Access to Octave 2 and EVI 4 mini is available on the platform, with plans for more languages and features in the future.