8 papers
Adaptation of Tacotron2-based Text-To-Speech for Articulatory-to-Acoustic Mapping using Ultrasound Tongue Imaging
Csaba Zainkó, László Tóth, Amin Honarmandi Shandiz +4
For articulatory-to-acoustic mapping, typically only limited parallel training data is available, making it impossible to apply fully end-to-end solutions like Tacotron2. In this p…
Speech Synthesis from Text and Ultrasound Tongue Image-based Articulatory Input
Tamás Gábor Csapó, László Tóth, Gábor Gosztolya +1
Articulatory information has been shown to be effective in improving the performance of HMM-based and DNN-based text-to-speech synthesis. Speech synthesis research focuses traditio…
Neural Speaker Embeddings for Ultrasound-based Silent Speech Interfaces
Amin Honarmandi Shandiz, László Tóth, Gábor Gosztolya +2
Articulatory-to-acoustic mapping seeks to reconstruct speech from a recording of the articulatory movements, for example, an ultrasound video. Just like speech signals, these recor…
Improving Neural Silent Speech Interface Models by Adversarial Training
Amin Honarmandi Shandiz, László Tóth, Gábor Gosztolya +2
Besides the well-known classification task, these days neural networks are frequently being applied to generate or transform data, such as images and audio signals. In such tasks,…
Applying Speech Tempo-Derived Features, BoAW and Fisher Vectors to Detect Elderly Emotion and Speech in Surgical Masks
Gábor Gosztolya, László Tóth
The 2020 INTERSPEECH Computational Paralinguistics Challenge (ComParE) consists of three Sub-Challenges, where the tasks are to identify the level of arousal and valence of elderly…
Ultrasound-based Articulatory-to-Acoustic Mapping with WaveGlow Speech Synthesis
Tamás Gábor Csapó, Csaba Zainkó, László Tóth +2
For articulatory-to-acoustic mapping using deep neural networks, typically spectral and excitation parameters of vocoders have been used as the training targets. However, vocoding…