Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285048116> ?p ?o ?g. }
Showing items 1 to 43 of
43
with 100 items per page.
- W4285048116 abstract "With the recent advance of neural networks in audio speech recognition (ASR), Deep Neural Network Based ASR has been widely used in multiple application scenarios such as smart homes, intelligent customer service, meeting minutes, and real-time translation. However, due to the ethnic variety of China, the pronunciation for the same word is different, which is a big challenge to the speech recognition system, especially for Short Utterances. This paper analyzes three kinds of commonly used speech feature parameters: spectrogram, MFCC (Mel-scale Frequency Cepstral Coefficients), and Fbank (Filter bank) and builds the dialect classification model based on a deep neural network for a dataset of 10 dialects in China. In detail, we study the geographical features of dialects and propose a multi-task model that uses the area of the dialect as an auxiliary task and builds a hard parameter sharing based multi-task learning model. The results show that the performance of this model can achieve up to 79.96%. Furthermore, as the hard parameter sharing model cannot effectively learn the correlation between sub-tasks, we then propose a sparse parameter sharing based multitask learning model. The model uses joint training to automatically learn the correlation between sub-tasks, prune redundant networks, and share network parameters. The experiment results show that the sparse parameter sharing for the multi-task classification model achieves the best accuracy, with an average of 83.59%." @default.
- W4285048116 created "2022-07-13" @default.
- W4285048116 creator A5023610301 @default.
- W4285048116 creator A5026144539 @default.
- W4285048116 creator A5050966388 @default.
- W4285048116 creator A5074461919 @default.
- W4285048116 creator A5080162995 @default.
- W4285048116 creator A5086744372 @default.
- W4285048116 date "2022-03-01" @default.
- W4285048116 modified "2023-09-24" @default.
- W4285048116 title "Deep Neural Network Based Chinese Dialect Classification" @default.
- W4285048116 doi "https://doi.org/10.1109/cbd54617.2021.00043" @default.
- W4285048116 hasPublicationYear "2022" @default.
- W4285048116 type Work @default.
- W4285048116 citedByCount "1" @default.
- W4285048116 countsByYear W42850481162023 @default.
- W4285048116 crossrefType "proceedings-article" @default.
- W4285048116 hasAuthorship W4285048116A5023610301 @default.
- W4285048116 hasAuthorship W4285048116A5026144539 @default.
- W4285048116 hasAuthorship W4285048116A5050966388 @default.
- W4285048116 hasAuthorship W4285048116A5074461919 @default.
- W4285048116 hasAuthorship W4285048116A5080162995 @default.
- W4285048116 hasAuthorship W4285048116A5086744372 @default.
- W4285048116 hasConcept C154945302 @default.
- W4285048116 hasConcept C204321447 @default.
- W4285048116 hasConcept C28490314 @default.
- W4285048116 hasConcept C2984842247 @default.
- W4285048116 hasConcept C41008148 @default.
- W4285048116 hasConcept C50644808 @default.
- W4285048116 hasConceptScore W4285048116C154945302 @default.
- W4285048116 hasConceptScore W4285048116C204321447 @default.
- W4285048116 hasConceptScore W4285048116C28490314 @default.
- W4285048116 hasConceptScore W4285048116C2984842247 @default.
- W4285048116 hasConceptScore W4285048116C41008148 @default.
- W4285048116 hasConceptScore W4285048116C50644808 @default.
- W4285048116 hasFunder F4320321001 @default.
- W4285048116 hasFunder F4320335787 @default.
- W4285048116 hasLocation W42850481161 @default.
- W4285048116 hasOpenAccess W4285048116 @default.
- W4285048116 hasPrimaryLocation W42850481161 @default.
- W4285048116 isParatext "false" @default.
- W4285048116 isRetracted "false" @default.
- W4285048116 workType "article" @default.