no code implementations • CL (ACL) 2022 • Veronika Vincze, Martina Katalin Szabó, Ildikó Hoffmann, László Tóth, Magdolna Pákáski, János Kálmán, Gábor Gosztolya
In order to ascertain the most useful features for distinguishing healthy controls, MCI patients, and mAD patients, we carry out a statistical analysis of the data and investigate the significance level of the extracted features among various speaker group pairs and for various speaking tasks.
no code implementations • 30 May 2023 • László Tóth, Amin Honarmandi Shandiz, Gábor Gosztolya, Csapó Tamás Gábor
Thanks to the latest deep learning algorithms, silent speech interfaces (SSI) are now able to synthesize intelligible speech from articulatory movement data under certain conditions.
1 code implementation • 26 Jul 2021 • Csaba Zainkó, László Tóth, Amin Honarmandi Shandiz, Gábor Gosztolya, Alexandra Markó, Géza Németh, Tamás Gábor Csapó
In this paper, we experimented with transfer learning and adaptation of a Tacotron2 text-to-speech model to improve the final synthesis quality of ultrasound-based articulatory-to-acoustic mapping with a limited database.
1 code implementation • 5 Jul 2021 • Tamás Gábor Csapó, László Tóth, Gábor Gosztolya, Alexandra Markó
Besides, we analyze the ultrasound tongue recordings of several speakers, and show that misalignments in the ultrasound transducer positioning can have a negative effect on the final synthesis performance.
no code implementations • 7 Aug 2020 • 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 speakers, determine whether the actual speaker wearing a surgical mask, and estimate the actual breathing of the speaker.
1 code implementation • 6 Aug 2020 • Tamás Gábor Csapó, Csaba Zainkó, László Tóth, Gábor Gosztolya, Alexandra Markó
The training target is the 80-dimensional mel-spectrogram, which results in a finer detailed spectral representation than the previously used 25-dimensional Mel-Generalized Cepstrum.
Audio and Speech Processing Sound
no code implementations • 24 Jun 2019 • Tamás Gábor Csapó, Mohammed Salah Al-Radhi, Géza Németh, Gábor Gosztolya, Tamás Grósz, László Tóth, Alexandra Markó
Recently it was shown that within the Silent Speech Interface (SSI) field, the prediction of F0 is possible from Ultrasound Tongue Images (UTI) as the articulatory input, using Deep Neural Networks for articulatory-to-acoustic mapping.
Sound Audio and Speech Processing
no code implementations • 11 Oct 2016 • Gábor Gosztolya, Tamás Grósz, László Tóth
Recently, attempts have been made to remove Gaussian mixture models (GMM) from the training process of deep neural network-based hidden Markov models (HMM/DNN).