Omilia has launched Lexis, a generative text-to-speech (TTS) model designed specifically for enterprise-grade contact centers.
Built natively into the Omilia Cloud Platform (OCP), Lexis aims to provide human-quality voice synthesis without the need for third-party API integrations, addressing common industry challenges regarding latency and data sovereignty.
By integrating the TTS engine directly into its stack, Omilia seeks to eliminate the bolted-on approach often found in conversational AI deployments. This native architecture ensures that audio data remains within the platform’s perimeter, providing brands with greater control over their vocal identity and reducing the technical friction associated with routing audio through external services.
Dimitris Vassos, CEO and Co-Founder at Omilia, said:
Dimitris Vassos, CEO and Co-Founder, Omilia
“The enterprise CX market is full of platforms that assemble voice from third-party APIs and call it integrated. We built Lexis natively into OCP because sovereignty, latency, and brand control cannot be delivered any other way. When an enterprise puts their voice in front of millions of customers, they need to own it — the sound, the data, and the guarantee that it will still be there tomorrow.”
Technical Capabilities and Performance
Lexis utilizes a generative deep neural network and NLP engine to analyze context, tone, and emotion at the sentence level. This allows the system to adjust pacing and intonation dynamically based on the nature of the customer interaction.
According to Omilia, the key capabilities of the Lexis engine include:
Sub-45ms Latency: Real-time streaming designed to facilitate natural, flowing conversations between AI agents and customers.
Custom Voice Cloning: The ability to create a deployable branded voice using a short, high-quality recording.
Pre-tuned CX Ambassador Personas: A library of production-ready voices tailored for specific roles, such as “The Trusted Advisor” or “The Quick Operator.”
Global Reach: Support for over 25 locales, featuring native pronunciation and prosody.
Granular Control: A two-layer tuning system that allows for custom pronunciation dictionaries and SSML support, alongside model-layer adjustments for pitch, warmth, and emotional range.
Data Security: A “zero data leakage” architecture where audio never leaves the OCP perimeter.
Claudio Rodriques, Chief Product Officer at Omilia, said:
Claudio Rodriques, Chief Product Officer, Omilia
“When TTS is bolted on from outside the platform, you inherit the latency of external round-trips, the compliance risk of audio leaving your perimeter, and the fragility of a voice library you do not own. Lexis solves all three at once — because it is not plugged into OCP, it is built into it.”
Compliance and Enterprise Readiness
To meet the rigorous demands of regulated industries, Lexis is certified under several compliance frameworks, including PCI-DSS, SOC 2 Type II, ISO 27001, HIPAA, GDPR, and CCPA. The engine is available as a fully managed SaaS offering or as an on-premise deployment for organizations with strict data-residency requirements.
By bringing TTS in-house, Omilia is addressing the latency issues that frequently lead to customer frustration and agent escalation. Furthermore, as data privacy regulations tighten globally, the ability to process voice synthesis within a single, secure perimeter—without third-party API exposure—provides a compelling blueprint for how financial and healthcare institutions can safely adopt generative AI in their contact centers.
Lexis is available now for all customers using the Omilia Conversational Platform.
