#1 Real-Time AI Voice SDK | Krisp
AI Meeting Assistant
with #1 Noise Cancellation
AI Notetaker
AI Note Taker](/content/ai-note-taker/index.html)
Meeting Transcription](/content/meeting-transcription/index.html)
Meeting Recording](/content/meeting-recording/index.html)
Meeting Summary](/content/ai-meeting-summary/index.html)
Real Time Voice AI
Noise Cancellation](/content/noise-cancellation/index.html)
Accent Conversion - Speaker side](/content/ai-accent-conversion/index.html)
Accent Conversion - Listener side](/content/ai-accent-conversion/listener/index.html)
Call Center AI
AI that boosts call center productivity
Speech Assist
Noise Cancellation](/content/contact-center/noise-cancellation/index.html)
Accent Conversion](/content/contact-center/accent-conversion/index.html)
Voice Translation](/content/ai-voice-translation/index.html)
Agent & Supervisor Assist
Agent Assist](/content/contact-center/ai-agent-assist/index.html)
Speech Analytics](/content/contact-center/speech-analytics/index.html)
Voice security](/content/contact-center/voice-security/index.html)
Developers
with #1 AI Voice Models
For Voice AI Agents
Voice Isolation](/content/developers/#voice-isolation/index.html)
Turn-Taking](/content/developers/#turn-taking/index.html)
For Human-to-human Calls
Accent Conversion](/content/developers/#accent-conversion/index.html)
Noise Cancellation](/content/developers/#noise-cancellation/index.html)
Voice Translation API](/content/developers/voice-translation-api/index.html)
Voice AI
Production-ready models that clean, understand, and transform audio in real time.
Krisp VIVA SDK
Voice infrastructure for voice AI agents.
Voice Isolation
Server-side voice isolation that sits in front of your VAD or STT. Removes background noise, cross-talk, and non-primary speakers before they reach your agent.
Turn Prediction
Predicts when a speaker is done — directly from raw audio, no transcription needed.
Interruption Prediction
Classifies user speech mid-response in real time — interruption vs. question.
Voice Activity Detection (VAD)
Accurate Voice Activity Detector — robust against background noise and secondary voices.
Krisp RTC SDK
Enhances communication in calls and meetings.
Accent Conversion
Converts a contact center agent's accent to match the customer's in real time.
Background Voice Cancellations
Removes background noise for the outbound/uplink stream.
Outbound Noise Cancellation
Removes other voices near the primary speaker.
Inbound Noise Cancellation
Strips background noise from mobile and PSTN callers.
Voice Translation
API for real-time speech-to-speech translation across 61 languages.
Frequently asked questions
- How does Voice Isolation differ from noise cancellation? Voice Isolation removes both background noise and secondary human voices.
- What's the difference between Turn Prediction and Interruption Prediction? Turn Prediction identifies when a speaker is about to finish talking, while Interruption Prediction identifies user intent.
- Do VIVA models require transcription? No, they operate directly on audio, are language agnostic, and support multiple languages natively.
- Can VIVA models be used together or independently? Each model can be deployed independently or in combination, depending on needs.
- What are the deployment requirements? VIVA models are optimized for on-server CPU deployment, integrating into existing voice pipelines.
- Can I use RTC and VIVA models together? Yes, deploy models from each family where needed based on use case.
- How many languages does Voice Translation support? Voice Translation supports 60+ languages for real-time bidirectional translation.