ПІДВИЩЕННЯ ІНФОРМАТИВНОСТІ ОСНОВНОГО ТОНУ ДЛЯ РОЗПІЗНАВАННІ МОВЦІВ ЗГОРТАЛЬНИМИ НЕЙРОМЕРЕЖАМИ
In the article the method and algorithm were presented for speaker’s pitch extraction from the speech signal, using the data of the wavelet transform and singular analysis of the speech signal. That allows arbitrary scale representation of the signal in the frequency domain and increases reliability...
Збережено в:
| Дата: | 2018 |
|---|---|
| Автори: | , |
| Формат: | Стаття |
| Мова: | Ukrainian |
| Опубліковано: |
Vinnytsia National Technical University
2018
|
| Теми: | |
| Онлайн доступ: | https://oeipt.vntu.edu.ua/index.php/oeipt/article/view/482 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Optoelectronic Information-Power Technologies |
Репозитарії
Optoelectronic Information-Power Technologies| Резюме: | In the article the method and algorithm were presented for speaker’s pitch extraction from the speech signal, using the data of the wavelet transform and singular analysis of the speech signal. That allows arbitrary scale representation of the signal in the frequency domain and increases reliability allocation pitch in the case of noise present in the speech signal and using a convolution neural network for speaker recognition. |
|---|