Fact · Speech recognition
0.5seconds (sampling interval)
The whisper.cpp stream example samples the microphone audio every half a second and runs the transcription continuously.
Checked against its source on October 17, 2026
In the source's own words
“The `whisper-stream` tool samples the audio every half a second and runs the transcription continuously.”
whisper.cpp stream example README, ggml-org, GitHub (ggml-org/whisper.cpp), n.d., Opening paragraph.
Open the original- What it measures
- A sample program in the whisper.cpp project, with the default step of 500 milliseconds.
- What it does not show
- The README calls it a naive example. Its results are sample output, not a benchmark of latency.