Conditional Spoken Digit Generation with StyleGAN

Kasperi Palkama, Lauri Juvela, Alexander Ilin

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussaConference article in proceedingsScientificvertaisarvioitu

5 Sitaatiot (Scopus)
193 Lataukset (Pure)

Abstrakti

This paper adapts a StyleGAN model for speech generation with minimal or no conditioning on text. StyleGAN is a multiscale convolutional GAN capable of hierarchically capturing data structure and latent variation on multiple spatial (or temporal) levels. The model has previously achieved impressive results on facial image generation, and it is appealing to audio applications due to similar multi-level structures present in the data. In this paper, we train a StyleGAN to generate melspectrograms on the Speech Commands dataset, which contains spoken digits uttered by multiple speakers in varying acoustic conditions. In a conditional setting our model is conditioned on the digit identity, while learning the remaining data variation remains an unsupervised task. We compare our model to the current unsupervised state-of-the-art speech synthesis GAN architecture, the WaveGAN, and show that the proposed model outperforms according to numerical measures and subjective evaluation by listening tests.

AlkuperäiskieliEnglanti
OtsikkoProceedings of Interspeech
KustantajaInternational Speech Communication Association (ISCA)
Sivut3166-3170
Sivumäärä5
DOI - pysyväislinkit
TilaJulkaistu - 2020
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInterspeech - Shanghai, Kiina
Kesto: 25 lokak. 202029 lokak. 2020
Konferenssinumero: 21
http://www.interspeech2020.org/

Julkaisusarja

NimiInterspeech
KustantajaInternational Speech Communication Association
ISSN (elektroninen)1990-9772

Conference

ConferenceInterspeech
LyhennettäINTERSPEECH
Maa/AlueKiina
KaupunkiShanghai
Ajanjakso25/10/202029/10/2020
www-osoite

Sormenjälki

Sukella tutkimusaiheisiin 'Conditional Spoken Digit Generation with StyleGAN'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä