#1 AI Voice Models
Low-latency, low-CPU

Voice Isolation for Voice AI Agents

Krisp VIVA family of models designed for server-side audio streams and integrated on-server in front of VAD or STT. Removes background noises and secondary voices for improved “turn detection” (aka “unwanted interruptions”) for Voice AI Agents and Bots. Language and accent independent.
Use Cases: Voice AI Agents

Turn-Taking for Voice AI Agents

Krisp VIVA family small model designed for seamless real-time conversations with Voice AI Agents. Detects when a speaker is likely to finish their turn, enabling Voice AI Agents to respond naturally, without awkward pauses or interruptions. Lightweight and optimized for on-server CPU deployment. Operates purely on audio input — no transcription required.
Use Cases: Voice AI Agents

Accent Conversion for Calls

Krisp RTC family model designed for changing the accent of a contact center agent in real-time from their local accent to that of the customer, improving customer experience while providing a significant ROI. Local accent robust, gender independent, noise robust and microphone/headset independent.
Use Cases: Soft Phones, CCaaS, Video meetings, Audio conferencing and streaming

Background Voice Cancellations for Calls

Krisp RTC family model designed (outbound/uplink) audio stream. Noise robust for use in contact centers, offices and home environments. Includes de-reverberation to remove room echo. Physical microphone and language independent.
Use Cases: Video meetings, Soft Phones, CCaaS, Audio conferencing and streaming

Outbound Noise Cancellation for Calls

Krisp RTC family model designed (outbound/uplink) audio stream to remove other voices near the primary speaker wearing a headset or ear buds. Fixes the contact center cross-talk problem, as well as blocking competing voices in office and at home. No enrollment or training of the user’s voice is required. Language independent.
Use Cases: Video meetings, Soft Phones, CCaaS, Audio conferencing and streaming

Inbound Noise Cancellation for Calls

Krisp RTC family model designed for the speaker (inbound/downlink) audio stream. Noise robust for CCaaS and UCaaS platforms so that customers calling in on mobile or PTSN have their background noise removed so that agents can have clear communications bi-directionally. Physical speaker or headset independent. Language independent. Intelligent pass-through of ringtones.
Use Cases: Video meetings and Soft Phones