Deepslate, a Brandenburg-based AI model lab, has raised €7.7 million in a Seed round to advance speech-to-speech AI models built and hosted entirely in Europe. The investment was led by 42CAP, with participation from Alstin Capital, existing investor SIVentures, formerly Smart Infrastructure Ventures, and several business angels. Headquartered in Templin and operating with between 11 and 50 employees, the company plans to use the new capital to strengthen research, data, infrastructure, commercial operations, and early production growth.
Funding Details
The funding round provides Deepslate with fresh capital to expand its research team and accelerate development of its native speech-to-speech models. The company disclosed that the round was led by 42CAP and included Alstin Capital and SIVentures, which was previously known as Smart Infrastructure Ventures. Several business angels also participated, and the company, whose leadership includes Paskal Paesler and Jan Brachthäuser, is actively hiring across open roles as it moves into its next phase of growth.
Direct Audio Processing
Deepslate develops end-to-end speech-to-speech models that process spoken language directly rather than converting speech into text before generating a response. The company says this approach reduces latency while retaining information such as tone, emphasis, dialect, and other aspects of spoken language. Its system consists of three components trained in-house: a speech encoder, a reasoning core based on an open-weights language model that Deepslate post-trains for specific European languages, and a speech decoder that produces the spoken reply.
Performance and Production Use
Deepslate reports that its model achieved a response time of 440 milliseconds in an independent Artificial Analysis benchmark, making it the fastest speech-to-speech model measured as of September 2026. The company also states that its technology posted the best error rate for European languages in the CoVoST2 comparison field and holds ISO 27001 certification. The technology is already used in production by insurers, contact centers, and platforms, with access via a self-service platform, an API, and volume or self-hosting options.
European Data Sovereignty
Deepslate positions European data sovereignty as a foundational requirement for its customers rather than an optional feature. The company discloses where computing takes place, who its subprocessors are, and which aspects remain open so that sovereignty claims can be verified rather than assumed. Its technology is hosted entirely within the EU, and customers can also run the models on their own infrastructure to meet stricter compliance needs or internal security policies.
Use of Funds
The new capital will support expanded model training and a European data program focused on German street names, personal names, and dialects while also reducing latency and improving voice quality. Deepslate plans to grow its sales and marketing teams and scale production infrastructure in European data centers to handle increasing call volumes. The company offers pricing from 2 cents per minute through its self-serve API and is building its go-to-market team to support commercial expansion.
Deepslate's Seed round highlights rising demand for voice AI that combines competitive speed and accuracy with verifiable European data sovereignty. With backing from 42CAP, Alstin Capital, SIVentures, and business angels, the company is positioned to expand research, data, and commercial operations. Its direct audio processing and EU hosting model may strengthen its role among insurers, contact centers, and platform providers seeking compliant voice automation.