Learning emotions latent representation with CVAE for Text-Driven Expressive AudioVisual Speech Synthesis

Autor: Slim Ouni, Valérian Girard, Vincent Colotte, Sara Dahmani

Publikováno v: Neural Networks
Neural Networks, Elsevier, 2021, 141, pp.315-329. ⟨10.1016/j.neunet.2021.04.021⟩
Neural Networks, 2021, 141, pp.315-329. ⟨10.1016/j.neunet.2021.04.021⟩

International audience; Great improvement has been made in the field of expressive audiovisual Text-to-Speech synthesis (EAVTTS) thanks to deep learning techniques. However, generating realistic speech is still an open issue and researchers in this a

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fb49718d2c957d8ff1ca49d1428e6a35
https://hal.inria.fr/hal-03204193

Transfer learning of the expressivity using flow metric learning in multispeaker text-to-speech synthesis

Autor: Ajinkya Kulkarni, Vincent Colotte, Denis Jouvet

Publikováno v: INTERSPEECH 2020
INTERSPEECH 2020, Oct 2020, Shanghai / Virtual, China
INTERSPEECH

International audience; In this paper, we present a novel deep metric learning architecture along with variational inference incorporated in a paramet-ric multispeaker expressive text-to-speech (TTS) system. We proposed inverse autoregressive flow (I

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::66f3de9ad2a382e71bb19836d45dc112
https://hal.inria.fr/hal-02572106v3

Deep Variational Metric Learning for Transfer of Expressivity in Multispeaker Text to Speech

Autor: Vincent Colotte, Ajinkya Kulkarni, Denis Jouvet

Publikováno v: Statistical Language and Speech Processing ISBN: 9783030594299
SLSP
SLSP 2020-8th International Conference on Statistical Language and Speech Processing
SLSP 2020-8th International Conference on Statistical Language and Speech Processing, Oct 2020, Cardiff / Virtual, United Kingdom

In this paper, we propose an approach relying on multiclass N-pair loss based deep metric learning in recurrent conditional variational autoencoder (RCVAE). We used RCVAE for implementation of multispeaker expressive text-to-speech (TTS) system. The

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::56935c6443b934c97f625ca3326142e6
https://doi.org/10.1007/978-3-030-59430-5_13

$$F_{0}$$ F 0 Modeling Using DNN for Arabic Parametric Speech Synthesis

Autor: Vincent Colotte, Denis Jouvet, Imene Zangar, Zied Mnasri

Publikováno v: Proceedings of the International Neural Networks Society ISBN: 9783030168407
INNSBDDL

Deep neural networks (DNN) are gaining increasing interest in speech processing applications, especially in text-to-speech synthesis. Actually state-of-the-art speech generation tools, like MERLIN and WAVENET are totally DNN-based. However, every lan

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_________::521a4d197ee8f01fc5f25aa4e8db21a5
https://doi.org/10.1007/978-3-030-16841-4_20

Duration modeling using DNN for Arabic speech synthesis

Autor: Denis Jouvet, Amal Houidhek, Zied Mnasri, Imene Zangar, Vincent Colotte

Publikováno v: 9th International Conference on Speech Prosody
9th International Conference on Speech Prosody, Jun 2018, Poznań, Poland

International audience; Duration modeling is a key task for every parametric speech synthesis system. Though such parametric systems have been adapted to many languages, no special attention was paid to explicitly handling Arabic speech characteristi

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ebfbda24a4a53b83db8a41e285f2ce56
https://hal.inria.fr/hal-01889917/document