Speaker-independent classification of phonetic segments from raw ultrasound in child speech improves when models receive the mean ultrasound frame as additional speaker information.
Com- parison of DCT and autoencoder-based features for DNN- HMM multimodal silent speech recognition,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Speaker-independent classification of phonetic segments from raw ultrasound in child speech
Speaker-independent classification of phonetic segments from raw ultrasound in child speech improves when models receive the mean ultrasound frame as additional speaker information.