Optimizing the Ultrasound Tongue Image Representation for Residual Network-Based Articulatory-to-Acoustic Mapping.
Tamás Gábor CsapóGábor GosztolyaLászló TóthAmin Honarmandi ShandizAlexandra MarkóPublished in: Sensors (2022)
Keyphrases
- image representation
- ultrasound images
- vocal tract
- multiscale
- image classification
- object recognition
- acoustic features
- image content
- bag of words
- quadtree
- image features
- feature space
- feature representations
- representation scheme
- image retrieval
- visual words
- receptive fields
- speech synthesis
- sparse coding
- image classification and retrieval
- scene recognition
- scene classification
- sparse representation
- object detection
- compressive sensing
- bag of features
- bag of visual words
- speech signal
- speech recognition
- scene categorization
- pattern recognition