Records found: 5
'StepFun AI released Step-Audio 2 Mini, an open-source 8B speech-to-speech model that combines unified audio-text tokenization, emotion-aware generation, and retrieval-augmented grounding to beat GPT-4o-Audio on multiple benchmarks.'
'Voice AI reached a turning point in 2025 as speech-native architectures, multimodal integration, and privacy-first on-device processing drove rapid enterprise and consumer adoption. Market leaders and startups are accelerating real-time, emotion-aware voice systems across healthcare, retail, and automotive.'
'OpenAI released GPT-Realtime and Realtime API with unified audio processing, SIP phone support and MCP server integration, improving performance and enterprise deployment options while key speech AI challenges remain.'
OpenAI has rolled out four significant updates to its AI agent framework, including a TypeScript SDK, RealtimeAgent for voice applications with human-in-the-loop control, enhanced tracing capabilities, and improvements to its speech-to-speech pipeline.
Chinese researchers release LLaMA-Omni2, a modular speech language model that enables real-time spoken dialogue with minimal latency and strong performance using compact training data.