Krisp Launches Real-Time Voice Translation SDK for Customer Conversations

Krisp, a startup working on real-time voice AI technology, has launched a new voice translation software development kit (SDK) that allows customer experience (CX) platforms to add real-time voice-to-voice translation directly into live customer conversations.

The company said the technology has already been deployed in real customer environments since 2025 through its Call Center AI platform. With the SDK launch, Krisp is now making the technology available to developers so they can integrate multilingual voice support into their own systems.

Unlike text translation tools, real-time voice translation must work instantly during live conversations. It needs to manage different accents, background noise and natural pauses between speakers while keeping delays very low. If translation is too fast, it may lose context and cause mistakes. If it is too slow, it can disrupt the flow of conversation.

Krisp said its SDK is designed to balance speed, accuracy, and natural conversation. It supports more than 60 languages and workers in two-way conversations, allowing customers and support agents to speak in different languages without needing a human interpreter.

“In global customer experience, every language barrier directly impacts speed and customer satisfaction,” said Davit Baghdasaryan, Co-founder and CEO of Krisp, in a statement. He added that making the technology available as an SDK will help companies reduce language friction and improve global customer support operations.

The SDK is available for Windows, macOS, and web applications. To improve accuracy, Krisp first applies local noise cancellation before processing audio in the cloud. The system also allows companies to add custom vocabulary and industry-specific terms to maintain consistency.

The Voice Translation SDK expands Krisp’s broader voice AI portfolio which includes noise cancellation, accent conversion, and voice isolation tools for voice AI agents.