Fact · Speech recognition
Microsoft defines the real-time factor (RTF) of a device as the ratio of its processing time to the audio length.
Checked against its source on October 17, 2026
In the source's own words
“the ratio of the processing time to the audio length”
Embedded speech performance evaluations (Speech service), Microsoft Learn, n.d., Metrics & terminology.
Open the original- What it measures
- Microsoft's guide to running speech models on a device, written for developers deciding whether a device is fast enough.
- What it does not show
- A metric for one device and one model. Microsoft says it should be measured from file-based input, not from live microphone input.