Dhvani adapts a weakly-supervised model to Hindi pronunciation error detection using synthetic mispronunciations, reporting 82% F1 on the synthetic test set but no validation on real non-native speech.
[5] The WEAKLY-S model, proposed by Korzekwa et al., used the robust S2S technique and bypassed the need for detailed phonetic transcription
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
method 1
citation-polarity summary
fields
eess.AS 1years
2025 1verdicts
REJECT 1roles
method 1polarities
support 1representative citing papers
citing papers explorer
-
Dhvani: A Weakly-supervised Phonemic Error Detection and Personalized Feedback System for Hindi
Dhvani adapts a weakly-supervised model to Hindi pronunciation error detection using synthetic mispronunciations, reporting 82% F1 on the synthetic test set but no validation on real non-native speech.