COMPARISON OF OPTIMIZATION ALGORITHMS OF CONNECTIONIST TEMPORAL CLASSIFIER FOR SPEECH RECOGNITION SYSTEM

Main Article Content

DOI

Yedilkhan Amirgaliyev

amir_ed@mail.ru

http://orcid.org/0000-0002-6528-0619
Kuanyshbay Kuanyshbay

darkhan.kuanyshbay@sdu.edu.kz

http://orcid.org/0000-0001-5952-8609
Aisultan Shoiynbek

aisultan.shoiynbek@sdu.edu.kz

http://orcid.org/0000-0002-9328-8300

Abstract

This paper evaluates and compares the performances of three well-known optimization algorithms (Adagrad, Adam, Momentum) for faster training the neural network of CTC algorithm for speech recognition. For CTC algorithms recurrent neural network has been used, specifically Long-Short-Term memory. LSTM is effective and often used model. Data has been downloaded from VCTK corpus of Edinburgh University. The results of optimization algorithms have been evaluated by the Label error rate and CTC loss.

Keywords:

recurrent neural network, search methods, acoustic, systems modeling language

References

Article Details

Amirgaliyev, Y., Kuanyshbay, K., & Shoiynbek, A. (2019). COMPARISON OF OPTIMIZATION ALGORITHMS OF CONNECTIONIST TEMPORAL CLASSIFIER FOR SPEECH RECOGNITION SYSTEM . Informatyka, Automatyka, Pomiary W Gospodarce I Ochronie Środowiska, 9(3), 54–57. https://doi.org/10.35784/iapgos.234