TI-ASU: Toward Robust Automatic Speech Understanding through Text-to-speech Imputation Against Missing Speech Modality.
Tiantian FengXuan ShiRahul GuptaShrikanth S. NarayananPublished in: CoRR (2024)
Keyphrases
- text to speech
- speech understanding
- speech synthesis
- speech recognition
- missing data
- missing values
- prosodic features
- speech recognizer
- text to speech synthesis
- programming tool
- noisy environments
- english text
- pattern recognition
- automatic speech recognition
- stochastic context free grammars
- computer vision
- multi modal
- bayesian networks
- word processing
- neural network