← all releases
Kyutai·Feb 12, 2026·v1.0·open source
Hibiki-Zero
Hibiki-Zero is an advanced model for simultaneous speech-to-speech translation that eliminates the need for word-level aligned data, simplifying the training pipeline and enabling seamless scaling to diverse languages. It achieves state-of-the-art performance in translation accuracy, latency, voice transfer, and naturalness across multiple tasks.
AudioTTSAGI
overview
Hibiki-Zero is an advanced model for simultaneous speech-to-speech translation that eliminates the need for word-level aligned data, simplifying the training pipeline and enabling seamless scaling to diverse languages. It achieves state-of-the-art performance in translation accuracy, latency, voice transfer, and naturalness across multiple tasks.
key features
- 01Simultaneous speech-to-speech translation without word-level aligned data
- 02State-of-the-art performance in translation accuracy and latency
- 03Preserves speaker identity and speech naturalness
- 04Adaptable to new languages with less than 1000 hours of speech data
- 05Provides model weights, inference code, and a 45-hour multilingual speech benchmark for evaluation
related productsbrowse all →
Agents UIopen sourceAgents UI is an open-source component library developed by LiveKit, designed to accelerate the creation of voice agent interfaces. Built with React and shadcn, it offers pre-built components for controlling input/output, managing sessions, rendering transcripts, and visualizing audio streams, enabling developers to build multi-modal, agentic applications on LiveKit's real-time platform.Fractalsopen sourceFractals is an open-source recursive agentic task orchestrator developed by TinyAGI. It enables the decomposition of high-level tasks into self-similar trees of executable subtasks, which are then executed by an agent swarm in isolated git worktrees.