less than 1 second
“The CVI now operates with less than 1 second end-to-end latency”
As published on cartesia.ai, October 2024. Captured by usedby on Oct 8, 2026.
What happened
Tavus uses Cartesia's Sonic model as the voice generation layer of its Conversational Video Interface, which lets AI digital twins hold real-time video conversations. It also uses Sonic's voice cloning and design features to create a distinct voice for each AI replica.
Summary written by usedby from the source page, in English. The figures are those of Cartesia and Tavus, not ours.
- sub-90ms“Cartesia’s Sonic model, with its consistent sub-90ms latency, has enabled Tavus to simplify their infrastructure while meeting their stringent performance requirements.”
Ultra-Low Latency: Sonic’s sub-90 ms latency enables Tavus to achieve end-to-end response times of less than 1 second, creating truly conversational experiences.
Cartesia’s Sonic model is a game-changer for our Conversational Video Interface. Its ultra-low latency of 90ms and high-quality voice generation have enabled us to create truly immersive real-time conversations with AI digital twins. The natural voices and voice design capabilities have elevated our product to new heights.
What the story claims, and what we checked
We compared the story with its live page on Oct 8, 2026.
- The figure: less than 1 secondCheckedPrinted word for word on the page, near the name of Tavus.
- The passage quoted aboveCheckedCopied word for word from the page, near the name of Tavus.
- The publication dateCheckedRead from the page’s own metadata, never guessed.
- Tavus uses CartesiaCheckedConfirmed line. Latest check across sources: Oct 8, 2026.
- The result itselfNot checkedWe quote it; we did not measure it.




