Inworld AI Blog· The Inworld team·· 4 d agoPickAI score66
Inworld releases SAVA voice activity and turn detection model
Introducing SAVA
AI summary
Inworld released SAVA, a semantic acoustic voice activity model that merges voice activity detection and turn detection into a single model and is already running in its voice products. The model outputs two probabilities every 40 ms, for speech and end of turn, has 25.5M parameters, and processes each 40 ms frame in under 3 ms on a single CPU thread of an Apple M3 Max.
Why it matters
The original article provides comparison data against Silero VAD and smart-turn on the same pipeline, along with 25.5M parameters and per-frame latency, so readers can judge whether it can replace their existing voice pipeline.
Source: Inworld AI Blog · inworld.ai