Fact · Voice and AI
The 2023 study describes an endpointer as a model that detects whether the user has finished speaking.
Checked against its source on October 17, 2026
In the source's own words
“detects whether the user has finished speaking”
From User Perceptions to Technical Improvement: Enabling People Who Stutter to Better Use Speech Recognition (full text), Colin Lea et al., Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (arXiv preprint), 2023, Section 2.1, Speech Recognition Systems.
Open the original (DOI 10.48550/arXiv.2302.09044)- What it measures
- The authors describe the pipeline of a voice assistant. The endpointer works while the user is speaking, alongside the speech recognizer.
- What it does not show
- A general architecture described in the paper's background, not the design of one vendor's product.