Sparrow-2 is Tavus's new conversational-understanding model — it reads a whole audio scene every 10ms to decide when to listen, wait, or speak, and it reasons over background noise instead of cancelling it. It is not an open-source noise-cancellation tool.
2026-09-09 · by Moe Ameen
Published 2026-09-09 · Updated
Short answer
Sparrow-2 is a proprietary turn-taking model from Tavus, announced August 27, 2026, that decides when an AI should listen, wait or speak in live conversations. It updates every 10ms and reads background noise instead of cancelling it. On TurnBench's public dev split, Tavus reports 92.4% end-of-turn recall and 97.4% interruption recall, the best of fourteen published systems. It is generally available through Tavus.
Key takeaways
On August 27, 2026, Tavus — the conversational-video AI company behind the Phoenix, Raven, and Sparrow model family — announced Sparrow-2, a real-time conversational-understanding model built to handle turn-taking in live voice and video conversations. It is live now as a general-availability model inside Tavus's stack, reachable through the Tavus API, the no-code PAL Maker, and the Conversational Video Interface (CVI) that powers Tavus's real-time avatars. There is an interactive demo at sparrow2.tavuslabs.org.
Worth clearing up front: Sparrow-2 is not an open-source noise-cancellation tool. It is a proprietary turn-taking model, and its approach to noise is the opposite of cancellation. Instead of stripping background sound out of the signal, Sparrow-2 treats the entire incoming audio stream as evidence — modeling tone, pauses, speaker identity, backchannels ("mm-hm"), interruptions, and ambient noise together to decide, roughly every 10 milliseconds, whether the person is done talking, thinking, or being interrupted by the room. Tavus frames this as a new take on the "cocktail party problem": understand the whole scene rather than scrub it.
Tavus positions Sparrow-2 as state-of-the-art on TurnBench, an independent, triple-annotated benchmark of real human conversation with a public leaderboard and a field of roughly fourteen systems. On the benchmark's dev split, Tavus reports the highest end-of-turn recall in the published field (92.4%, against 84.1% for the runner-up) and the highest interruption recall (97.4%). It also says Sparrow-2 leaves 7.6% of turn ends unanswered, against about 1 in 6 for VAP, the strongest research baseline. These are preliminary results on the benchmark's public dev split.
The model runs at a native 10ms frame rate (up from Sparrow-1's 40ms) for faster, finer reactions. Treat the exact figures as Tavus-reported until independently reproduced, but the direction — sharper, faster turn detection in noisy, multi-speaker settings — is the story.
There are two honest ways to act on this, and neither is "use Sparrow-2 to make posts" — it does not make content. The first is coverage. "Sparrow-2" and "Tavus turn-taking" are high-intent searches this week while builders work out what the model actually is and how it differs from noise cancellation. A single clear point of view turns into a week of content inside Kompozy: a short explainer, a Carousel breaking down turn-taking vs. noise cancellation, captioned Clipped Shorts from a screen-recorded demo, a Blog Article, and platform-native Text Posts — all held to one voice by the Persona Brief and pushed across eight social platforms plus blog and email on autopilot while interest is still climbing.
The second is the workflow distinction the launch draws so cleanly. Tavus and Sparrow-2 own the live, two-way conversation — an avatar that listens and responds in real time. Kompozy owns the other half: producing scripted, on-brand avatar and short-form content you publish one-to-many. If you demo a Sparrow-2-powered agent on a call or a stream, Kompozy clips and captions that recording into shorts; and for net-new marketing you never have to film, Persona Shorts generate a talking-head avatar reading your script. Different problems, complementary tools — see the honest breakdown in our AI video generators compared, which already covers where Tavus wins.
This AI avatar video is a real output from the Kompozy workspace. It is scripted and published one-to-many, unlike a live Tavus conversation.
Best for:
Not ideal for:
No. Sparrow-2 is a proprietary turn-taking model from Tavus. Instead of removing background noise, it classifies speech, noise and other sounds and uses them to decide when an AI should talk.
It handles turn-taking in live voice and video conversations. It updates up to 100 times a second to judge whether a person has finished talking, is pausing or is interrupting.
On TurnBench's public dev split, Tavus reports 92.4% end-of-turn recall and 97.4% interruption recall, leading all fourteen published systems. Tavus calls these preliminary and says test-set results will follow.
No. It is infrastructure for live conversational agents. To produce and publish avatar video, clips, carousels or blogs, you would use a content engine like Kompozy.
We checked every figure, date, quote and product claim on this page against the sources listed below on 2026-10-23, and removed or rewrote anything we could not confirm. All benchmark figures and specs come from Tavus's own research post. We replaced a failure-rate claim we could not find there with the figures Tavus published. Kompozy plan and format details come from kompozy.io/pricing.
Sources