DeepL Expands Into Voice Translation With Real-Time Multilingual Suite

DeepL has introduced a real-time voice-to-voice translation suite for meetings, mobile and web conversations, and group sessions, reports TechCrunch. The system supports live translation during calls on platforms like Zoom and Microsoft Teams, allowing participants to hear or read translations in real time.

The suite also includes a mobile and web product for in-person or remote discussions, as well as group conversation tools accessible via QR codes. DeepL said its technology adapts to custom terminology, including industry-specific vocabulary and personal names.

The company is launching an API that enables developers and businesses to integrate DeepL’s voice translation capabilities into their own applications, such as call centers. The program is currently in early access, with organizations invited to join a waitlist.

DeepL stated that it currently processes speech by converting it to text, translating it, and then generating audio output, while working toward an end-to-end model that skips the text step. The company said it controls the entire translation stack and aims to reduce latency while maintaining accuracy.