Turn-detection simulator — VAD ➜ Smart Turn

Replays your audio through the exact Silero VAD state machine and Smart Turn buffering used in a live voice pipeline, independently for all three models. When VAD detects the user stopped speaking, the buffered segment is sent to Smart Turn and each model's decision is shown below, alongside the audio it was given.

Trained — int8 (cpu, 8.8 MB)

Trained — fp32 (gpu, 32 MB)

Baseline — smart-turn-v3.2-cpu