Zobrazeno 1 - 10
of 59
pro vyhledávání: '"Colotte Vincent"'
Flow-based generative models are widely used in text-to-speech (TTS) systems to learn the distribution of audio features (e.g., Mel-spectrograms) given the input tokens and to sample from this distribution to generate diverse utterances. However, in
Externí odkaz:
http://arxiv.org/abs/2305.17724
Autor:
Fauth Camille, Bonneau Anne, Mella Odile, Colotte Vincent, Fohr Dominique, Jouvet Denis, Laprie Yves, Trouvain Jürgen
Publikováno v:
SHS Web of Conferences, Vol 8, Pp 1237-1250 (2014)
La plupart des corpus en langue se focalisent sur les phénomènes linguistiques écrits et concernent l’anglais (voir le site web : « Learner corpora around the world » de l’Université de Louvain - Belgique). La recherche phonétique sur l’
Externí odkaz:
https://doaj.org/article/a60672fd0c724d42bbc21d99fc0a241b
Publikováno v:
SHS Web of Conferences, Vol 1, Pp 1471-1485 (2012)
Continuations intra et interphrastiques du français : premiers résultats expérimentaux Cet article rend compte d’un certain nombre d’observations pour l’étude des continuations mineures et majeures en français. Ces observations ont été o
Externí odkaz:
https://doaj.org/article/6aaf10ff58b140f0a434e9e03fcd62d4
Training of multi-speaker text-to-speech (TTS) systems relies on curated datasets based on high-quality recordings or audiobooks. Such datasets often lack speaker diversity and are expensive to collect. As an alternative, recent studies have leverage
Externí odkaz:
http://arxiv.org/abs/2210.06370
Publikováno v:
Language Resources and Evaluation, 2020 Oct 01. 54(4), 943-974.
Externí odkaz:
https://www.jstor.org/stable/48740918
Publikováno v:
In Neural Networks September 2021 141:315-329
Publikováno v:
INTERSPEECH 2022
INTERSPEECH 2022, Sep 2022, Incheon, South Korea
INTERSPEECH 2022, Sep 2022, Incheon, South Korea
International audience; The main objective of this work is to study the expressivity transfer in a speaker's voice for which no expressive speech data is available in non-autoregressive end-to-end TTS systems. We investigated the expressivity transfe
Publikováno v:
EUSIPCO 2022
EUSIPCO 2022, Aug 2022, Belgrade, Serbia
EUSIPCO 2022, Aug 2022, Belgrade, Serbia
International audience; The main goal of this work is to provide fine-grained transfer of expressivity in various speaker's voices for which no expressive speech data is available. Our approach conditions a multispeaker Tacotron 2 system with latent
Publikováno v:
The 2022 IEEE Spoken Language Technology Workshop (SLT 2022)
The 2022 IEEE Spoken Language Technology Workshop (SLT 2022), Jan 2023, Doha, Qatar
The 2022 IEEE Spoken Language Technology Workshop (SLT 2022), Jan 2023, Doha, Qatar
Training of multi-speaker text-to-speech (TTS) systems relies on curated datasets based on high-quality recordings or audiobooks. Such datasets often lack speaker diversity and are expensive to collect. As an alternative, recent studies have leverage
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::389eee725e0f93589d0c975877db7c64
https://hal.science/hal-03812715
https://hal.science/hal-03812715
Autor:
Caillat, Domitille, Marin, Ludovic, Dodane, Christelle, Hirsch, Fabrice, Ouni, Slim, Slangen, Pierre, Guyot, Patrice, Colotte, Vincent, Morgenstern, Aliyah, Abel, Louis, Grondin-Verdon, Mickaëlla, Goupil, Juliette Lozano
Publikováno v:
ISGS 2022-9th Conference of the International Society for Gesture Studies
ISGS 2022-9th Conference of the International Society for Gesture Studies, Jul 2022, Chicago, United States
ISGS 2022-9th Conference of the International Society for Gesture Studies, Jul 2022, Chicago, United States
International audience
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od_______165::aa1193e294f8dc2293234afb67831ecd
https://imt-mines-ales.hal.science/hal-03875218
https://imt-mines-ales.hal.science/hal-03875218