Improving Low Resource Turkish Speech Recognition with Data Augmentation and TTS

Ramazan Gokay, Hulya Yalcin

Araştırma sonucu: ???type-name???Konferans katkısıbilirkişi

19 Atıf (Scopus)

Özet

One of the major problems faced by speech recognition researchers is the lack of data. In this paper, our objective is to compare alternative solutions to lack of data. Some experiments are conducted with very limited training data to see the effects of data augmentation and speech synthesis on speech recognition. Speed and volume perturbations are applied in this study. Besides data augmentation, synthetic speech is generated by using two different speech synthesis methods. In first speech synthesis approach, Google Translate Text to Speech (gTTS) is used as speech synthesizer. In second speech synthesis approach, an end-to-end Turkish TTS system is trained by us. Finally, we examined the effects of all these alternative methods on speech recognition for low resource languages. Our results demonstrate that some data augmentation or speech synthesis techniques work well to improve speech recognition for low resource languages. In this study, 14.8% relative Word Error Ratio (WER) improvement is obtained by using combination of augmented and synthetic data.

Orijinal dilİngilizce
Ana bilgisayar yayını başlığı16th International Multi-Conference on Systems, Signals and Devices, SSD 2019
YayınlayanInstitute of Electrical and Electronics Engineers Inc.
Sayfalar357-360
Sayfa sayısı4
ISBN (Elektronik)9781728118208
DOI'lar
Yayın durumuYayınlandı - Mar 2019
Etkinlik16th International Multi-Conference on Systems, Signals and Devices, SSD 2019 - Istanbul, Turkey
Süre: 21 Mar 201924 Mar 2019

Yayın serisi

Adı16th International Multi-Conference on Systems, Signals and Devices, SSD 2019

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???16th International Multi-Conference on Systems, Signals and Devices, SSD 2019
Ülke/BölgeTurkey
ŞehirIstanbul
Periyot21/03/1924/03/19

Bibliyografik not

Publisher Copyright:
© 2019 IEEE.

Parmak izi

Improving Low Resource Turkish Speech Recognition with Data Augmentation and TTS' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Alıntı Yap