Kugel 3.5
Text-to-speech that beats any other Model out there
Real-time text-to-speech API with voice cloning and sub-50ms latency, hosted in Europe and GDPR compliant.
KugelAudio homepage
1 / 3Overview
What is Kugel 3.5?
Text-to-speech that beats any other Model out there
Kugel 3.5 is KugelAudio's newest real-time text-to-speech model, built for developers who need natural, fast voice output. It streams first audio in under 50 milliseconds, pronounces numbers, codes, phone numbers and short replies accurately, and offers voice cloning plus 100+ ready-to-use recorded voices across 39 languages with native accents. The models are developed and hosted in Europe with GDPR compliance, and on-premise deployment is available for enterprise customers. Access is through a WebSocket streaming API with Python, TypeScript and cURL SDKs, plus Pipecat and LiveKit integrations. Pricing is simple pay-as-you-go per audio minute, billed by the second — start free, no credit card required.
Platforms and languages
Kugel 3.5 Availability
Platforms
Languages
Capabilities
Kugel 3.5 Key Features
Real-time streaming TTS
First audio streams in under 50 milliseconds over WebSocket, fast enough for real-time voice agents and conversational products.
Voice cloning and voice library
Clone a voice in under a minute, or choose from 100+ ready-to-use recorded voices covering a wide range of styles.
Grammar-aware pronunciation
Accurately speaks numbers, IBANs, phone numbers, postal codes and email addresses, plus short replies, pauses and spelled-out codes.
39 languages, EU-hosted and GDPR compliant
39 languages with native accents and dialects; developed and hosted in Europe with GDPR compliance and EU Sovereign Cloud on every plan.
On-premise deployment and developer SDKs
Run on your own infrastructure with the same SDK and API as the hosted service; Python, TypeScript and cURL SDKs with Pipecat and LiveKit integrations.
Best for
Who uses Kugel 3.5?
Developers
building real-time voice products, voice assistants and AI agents with fast natural speech
Plans and access
Kugel 3.5 Pricing
Freemium
Starts free, no card required; pay-as-you-go from €0.035 per audio minute, billed by the second
Common questions
Kugel 3.5 FAQs
Is KugelAudio free to try?
Yes. You can start for free with no credit card required, then pay per audio minute, billed by the second.
How is usage priced?
Pricing is per audio minute, billed by the second: kugel-3.5 costs €0.0525/min, kugel-2-turbo €0.035/min, and kugel-3 €0.07/min (excl. VAT). Enterprise plans with on-premise and dedicated deployment are priced on request.
What languages and voices are supported?
39 languages with native accents and dialects. You can also clone a voice in under a minute, or pick from 100+ ready-to-use recorded voices.
Where is my data processed, and can I self-host?
Models are developed and hosted in Europe and are GDPR compliant. EU Sovereign Cloud is included on every plan for EU/EEA customers, and enterprise customers can deploy fully on-premise.
Reviews
See what the community thinks and share your experience.
5.0
Based on 1 ratings
Rating distribution
Community reviews
No reviews
No written reviews yet.
