Sparrow 2
Sparrow 2 is an audio native, streaming first, multilingual conversational understanding model. It jointly models semantics, prosody, speaker identity, backchannels, interruptions, and the full acoustic scene rather than relying on silence timers or noise cancellation, enabling more accurate turn taking and natural conversation flow in noisy or shared environments.
- Type
- Audio
- Weights
- Closed
Related tools
1
More from Tavus
11
-
Phoenix 1
Video
Phoenix-1 is Tavus’ early Phoenix research model for realistic avatar generation, using 3D reconstruction and…
-
Phoenix 2
Video
Phoenix-2 is Tavus’ next Phoenix research model that shifts the avatar rendering backbone toward 3D Gaussian…
-
Phoenix 3
Multimodal
Phoenix-3 is Tavus’ full-face real-time rendering model for hyper-realistic conversational avatars with persi…
-
Phoenix 4
Multimodal
Phoenix-4 is Tavus’ real-time human rendering model that generates continuous facial behavior with controllab…
-
Phoenix-4.5
Video
Phoenix-4.5 is Tavus's real-time human rendering model for AI video characters, generating full-frame video w…
-
Raven 0
Multimodal
Raven-0 is Tavus’ real-time perception model that continuously interprets what it sees to add visual context…
-
Sparrow 0
Multimodal
Sparrow-0 is Tavus’ conversational timing model that manages turn-taking so voice and video agents speak, pau…
-
Sparrow 1
Audio
Sparrow-1 is Tavus’s multilingual audio-native model for conversational timing, predicting who holds the floo…
About Tavus
1
Tools using Sparrow 2
9