Whispered Speech Recognition Based on Audio Data Augmentation and Inverse Filtering

Modern Automatic Speech Recognition (ASR) systems are primarily designed to recognize normal speech. Due to a considerable acoustic mismatch between normal speech and whisper, ASR systems suffer from a significant loss of performance in whisper recognition. Creating large databases of whispered spee...

Descrizione completa

Dettagli Bibliografici

Autori principali:	Jovan Galić, Branko Marković, Đorđe Grozdić, Branislav Popović, Slavko Šajić
Natura:	Artigo
Lingua:	Inglês
Pubblicazione:	MDPI AG 2024-09-01
Serie:	Applied Sciences
Soggetti:	artificial neural networks audio databases automatic speech recognition convolutional neural network hidden Markov models inverse filtering
Accesso online:	https://www.mdpi.com/2076-3417/14/18/8223
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!