Autoencoder-Based Articulatory-to-Acoustic Mapping for Ultrasound Silent Speech Interfaces.
Gábor GosztolyaÁdám PintérLászló TóthTamás GrószAlexandra MarkóTamás Gábor CsapóPublished in: IJCNN (2019)
Keyphrases
- acoustic features
- speech recognition
- vocal tract
- formant frequencies
- speech signal
- speech sounds
- speech recognition systems
- speech synthesis
- automatic speech recognition
- speaker independent
- speaker verification
- vowel phonemes
- multi stream
- noisy speech
- speech recognizers
- prosodic features
- multimodal interfaces
- visual features
- emotional speech
- music information retrieval
- spoken dialogue systems
- ultrasound images
- acoustic models
- noisy environments
- pattern recognition
- user interface
- hidden markov models
- audio stream
- text to speech
- underwater acoustic
- spoken language
- language model
- neural network
- speaker recognition
- acoustic signal
- mel frequency cepstral coefficients
- source localization
- speaker identification