The Speech Team within the Siri organization drives major speech recognition, synthesis and speech to speech model changes for various features deeply embedded throughout Apple’s ecosystem. Our mission is to build cutting-edge infrastructure, datasets, and models that empower Siri conversational AI, dictation and various speech enabled Apple Intelligence features with powerful capabilities across natural language understanding, dialog generation, speech recognition, and multi-modal interaction. We apply these technologies to create engaging, intelligent, and personalized conversational experiences for millions of Apple users. We believe that the most impactful breakthroughs in deep learning emerge when we address real-world problems at scale. We develop speech to speech experiences and the underlying multimodal foundation model technology for current and future speech-enabled features across Apple’s software, hardware, and services ecosystem. This allows for cutting edge applied research anchored in Apple specific production needs, while improving speech interaction experiences for Apple’s customers around the world.