MAE-AST: Masked Autoencoding Audio Spectrogram Transformer.
Alan BaadePuyuan PengDavid HarwathPublished in: CoRR (2022)
Keyphrases
- multimedia
- fuzzy logic
- audio video
- visual information
- speech signal
- pattern analysis
- digital video
- audio signals
- audio visual
- fault diagnosis
- correlation coefficient
- audio stream
- signal processing
- power transformers
- power system
- wigner distribution
- artificial intelligence
- emotion recognition
- cross modal
- high voltage
- speaker identification
- sound theoretical
- energy distribution
- feature extraction
- multiresolution
- stack filters
- distribution network
- audio signal
- audio features
- multi modal
- boolean functions
- speech recognition