HMamba, a hierarchical Mamba-based model with a decoupled cross-entropy loss, jointly performs pronunciation scoring and mispronunciation detection, reaching an MDD F1 of 63.85% on speechocean762.
These modules collectively generate the corresponding aspect score sequence 饾惉饾憯 for each linguistic granularity 饾憯, as well as the phonetic error states 饾悶 and diagnosis 饾惒
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards Efficient and Multifaceted Computer-assisted Pronunciation Training Leveraging Hierarchical Selective State Space Model and Decoupled Cross-entropy Loss
HMamba, a hierarchical Mamba-based model with a decoupled cross-entropy loss, jointly performs pronunciation scoring and mispronunciation detection, reaching an MDD F1 of 63.85% on speechocean762.