Fact · Speech recognition
150milliseconds (mouth-to-ear delay)
ITU-T Recommendation G.114 says that if mouth-to-ear speech delay can be kept below 150 ms, most speech and non-speech applications experience essentially transparent interactivity.
Checked against its source on October 17, 2026
In the source's own words
“if delays can be kept below this figure, most applications, both speech and non-speech, will experience essentially transparent interactivity”
ITU-T Recommendation G.114: One-way transmission time, International Telecommunication Union, ITU-T, 2003, Introduction to the delay limits.
Open the original- What it measures
- The recommendation is about telephone and VoIP calls, where a person hears their own voice and the other side's voice with a delay.
- What it does not show
- It describes conversation between people, not the wait for a transcript, and the recommendation itself warns that a few applications are affected below 150 ms.