Fact · Dictation
The 2019 paper proposes and tests a punctuation model built on prosody, the pitch and timing cues of speech, rather than on the words alone.
Checked against its source on October 16, 2026
In the source's own words
“proposes and evaluates a prosody inspired approach and a phrase sequence model”
An Audio-based Sequential Punctuation Model for ASR and its Effect on Human Readability, Gyorgy Szaszak, Acta Polytechnica Hungarica, 2019, Abstract.
Open the original- What it measures
- Abstract of a 2019 journal paper on punctuation in speech recognition output.
- What it does not show
- Tested on Hungarian read speech. The paper does not say that commercial dictation apps use this method.