Fact · Dictation software
In a 2025 study of speech recognition models on children's speech, a child's age and the number of words in the audio had the highest impact on accuracy, followed by background noise and pronunciation ability.
Checked against its source on October 16, 2026
In the source's own words
“physiology (age) and particular external factor (number of words in audio) have the highest impact on accuracy, followed by background noise and pronunciation ability”
Causal Analysis of ASR Errors for Children: Quantifying the Impact of Physiological, Cognitive, and Extrinsic Factors, Vishwanath Pratap Singh et al., arXiv, 2025, Abstract.
Open the original- What it measures
- The authors ran Wav2Vec2.0, HuBERT, Whisper and MMS across age groups on two children's speech corpora, then used causal inference to estimate how much each factor contributes to the error rate.
- What it does not show
- A research analysis of open models on two corpora, not a rating of any dictation app. The finding that age matters most is about how the models perform on child speech in general, so test any tool with the child's own voice.