KugelAudio

Active

Kugel is an on-prem enterprise TTS supporting 25+ languages

Spring 2026Founded 20254 peopleBerlin, Berlin, Germany

kugelaudio.com/en ↗LinkedIn ↗X ↗GitHub ↗Crunchbase ↗See on the Idea Map B2B momentum

Generate ideas →

AI insightcan contain mistakes

Multilingual Text-to-SpeechAPI/InfraEnterprise applications and content platformsHigh competition

Moat

On-premises deployment and multilingual quality are defensible if latency and cost advantages persist.

Key risk

Cloud TTS providers improving multilingual support and offering on-prem options could commoditize.

Why now

Global enterprises need on-prem TTS for privacy and latency-sensitive applications.

Competitors

Google Cloud TTS, Azure Speech, Elevenlabs

About

Kugel is an enterprise-ready TTS model, available on-prem, with a focus on 25+ languages and low latency.

From their website

as of Jun 7, 2026kugelaudio.com/en ↗

API/InfraUsage-based · Kugel Classic €0.0860 per generated minute; Kugel Turbo €0.0430 per generated minute. Token cost estimates: ~€61 / 1M tokens for Classic, ~€31 / 1M tokens for Turbo. Enterprise/custom API pricing with quotes for committed usage and on-premise options.

KugelAudio offers on-prem enterprise TTS hosted in Europe with 25+ (40+) languages, targeting GDPR-compliant, low-latency voice AI for business use. It provides European data sovereignty, on-prem options, and dedicated enterprise support. Pricing is API-based with classed options for Classic and Turbo models and enterprise custom pricing.

KugelAudio provides enterprise text-to-speech deployed in Europe with 40+ languages and ultra-low latency. It offers two API models: Kugel Classic (higher quality TTS) and Kugel Turbo (lower latency for streaming, IVR, and high concurrency). Pricing is per generated audio minute, with token costs estimated from minutes. Enterprise pricing includes custom API pricing, on-premise hosting options in selected regions (EU-only hosting), dedicated support, and scalable concurrency. The product supports production endpoints and streaming generation, with easy integration (two lines of code examples shown) and on-premise deployment tailored to enterprise needs.

Who it’s for: Enterprises and organizations requiring EU-hosted, GDPR-compliant voice AI with data sovereignty and on-prem deployment.

Features

40+ languages supported
Ultra-low latency (39 ms inference time to first audio)
On-premise and EU-only hosting options
Streaming generation for high-concurrency use
Dedicated enterprise support with priority response
Custom integrations and configurable deployments
Production-ready endpoints for voice AI

Pricing page indicates active product with API offerings and enterprise customization; mentions dedicated support and on-premise deployment, suggesting market-facing traction and enterprise adoption potential.

Founders · 2

Kajo KratzensteinFounder

Founder at KugelAudio (YC P26). Previously CTO at Sagemode, scaling an in-game AI coach to 1,000+ users with deep League of Legends integration. Built production aneurysm detection model at Floy GmbH. MA and BSc in Computer Science from Hasso Plattner Institute.

LinkedIn ↗

Viktor PresberCo-Founder

Co-Founder of KugelAudio, building foundational Text-To-Speech models for customer support. Started founding ventures at 17 with a local video production company. Later co-founded full-house.io, scaling to $300K ARR in six months with enterprise clients like Siemens Energy, Hitachi Rail, and Barilla. Previously co-founded Sagemode, securing partnerships with two major esports teams.

LinkedIn ↗

Launch · 1 launch

KugelAudio: A multilingual voice AI model you can run in your own cluster. · May 2026
View ↗
Human-like TTS for 30+ languages, on-prem and low latency
KugelAudio is a multilingual voice AI model generating natural speech in 30+ languages and dialects that runs fully on-premise in Kubernetes clusters with 60ms latency, supporting IPA notation and voice cloning.
▲ 7