AI BriefingKugelAudio Launches Low-Latency Real-Time TTS Model
·2026.05.28 05:30
Key point
A real-time TTS and voice cloning model supporting sub-60ms low latency and self-hosting has been released.
Details
KugelAudio, a real-time Text-to-Speech (TTS) model designed for voice agent optimization, has been launched. Its core feature is voice cloning, which can replicate a voice from just 30-60 seconds of audio sample.
Key technical features are as follows:
- Ultra-low latency performance: Achieves latency of under 60ms excluding network delay, optimized for real-time conversational AI.
- Flexible deployment options: In addition to API-based calls, it supports on-premise and self-hosting that can run on your own cluster for environments where data security is important.
- Advanced language processing: Supports 25+ languages and features grammar-aware normalization that naturally reads phone numbers, addresses, drug names, and more.
- Ecosystem integration: Provides adapters for LiveKit, Pipecat, Vapi, and enhances developer accessibility through Python, JS, and Java SDKs.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Foremerge v0.5.0 Released: Rust-Based Open-Source Tool Detects Intent Conflicts Between Parallel AI Coding Agents