DATA AUGMENTATION STRATEGIES FOR NEURAL NETWORK F0 ESTIMATION

Visa fullständig post



Permalänk

http://hdl.handle.net/10138/306497

Citation

Airaksinen , M , Juvela , L , Alku , P & Rasanen , O 2019 , DATA AUGMENTATION STRATEGIES FOR NEURAL NETWORK F0 ESTIMATION . in 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP) . International Conference on Acoustics Speech and Signal Processing ICASSP , IEEE , pp. 6485-6489 , 44th IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) , Brighton , 12/05/2019 . https://doi.org/10.1109/icassp.2019.8683041

Titel: DATA AUGMENTATION STRATEGIES FOR NEURAL NETWORK F0 ESTIMATION
Författare: Airaksinen, Manu; Juvela, Lauri; Alku, Paavo; Rasanen, Okko
Upphovmannens organisation: Department of Neurosciences
Kliinisen neurofysiologian yksikkö
University of Helsinki
HUS Neurocenter
Utgivare: IEEE
Datum: 2019
Språk: eng
Sidantal: 5
Tillhör serie: 2019 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP)
Tillhör serie: International Conference on Acoustics Speech and Signal Processing ICASSP
ISBN: 978-1-4799-8131-1
ISSN: 1520-6149
DOI: https://doi.org/10.1109/icassp.2019.8683041
Permanenta länken (URI): http://hdl.handle.net/10138/306497
Abstrakt: This study explores various speech data augmentation methods for the task of noise-robust fundamental frequency (F0) estimation with neural networks. The explored augmentation strategies are split into additive noise and channel-based augmentation and into vocoder-based augmentation methods. In vocoder-based augmentation, a glottal vocoder is used to enhance the accuracy of ground truth F0 used for training of the neural network, as well as to expand the training data diversity in terms of F0 patterns and vocal tract lengths of the talkers. Evaluations on the PTDB-TUG corpus indicate that noise and channel augmentation can be used to greatly increase the noise robustness of trained models, and that vocoder-based ground truth enhancement further increases model performance. For smaller datasets, vocoder-based diversity augmentation can also be used to increase performance. The best-performing proposed method greatly outperformed the compared F0 estimation methods in terms of noise robustness.
Subject: Speech analysis
F0 estimation
noise robustness
data augmentation
deep learning
SPEECH RECOGNITION
3124 Neurology and psychiatry
6121 Languages
Referentgranskad: Ja
Användningsbegränsning: openAccess
Parallelpublicerad version: publishedVersion


Filer under denna titel

Totalt antal nerladdningar: Laddar...

Filer Storlek Format Granska
untitled.pdf 263.6Kb PDF Granska/Öppna

Detta dokument registreras i samling:

Visa fullständig post