Wyniki wyszukiwania - BazTech

1

Poprawa obiektywnych wskaźników jakości mowy w warunkach hałasu

Kąkol K., Kostek B.

Zeszyty Naukowe Wydziału Elektrotechniki i Automatyki Politechniki Gdańskiej

|

2018

|

Nr 60

45--50

PL

Celem pracy jest modyfikacja sygnału mowy, aby uzyskać zwiększenie poprawy obiektywnych wskaźników jakości mowy po zmiksowaniu sygnału użytecznego z szumem bądź z sygnałem zakłócającym. Wykonane modyfikacje sygnału bazują na cechach mowy lombardzkiej, a w szczególności na efekcie podniesienia częstotliwości podstawowej F0. Sesja nagraniowa obejmowała zestawy słów i zdań w języku polskim, nagrane w warunkach ciszy, jak również w obecności sygnałów zakłócających, tj. szumu różowego oraz tzw. gwaru (ang. babble speech), określanego też jako efekt „cocktail-party”. W ramach badań przetwarzano próbki mowy głosów męskich. W pracy wykazano, że podniesienie częstotliwości podstawowej skutkuje zwiększonymi wartościami wskaźnika jakości mowy, mierzonymi przy użyciu standardu PESQ (Perceptual Evaluation of Speech Quality).

EN

The aim of the work is to modify the speech signal in order to improve objective speech quality indicators after mixing the useful signal with noise or with an interfering signal. Modifications made to the signal are based on the characteristics of the Lombard speech, and in particular on the effect of raising the fundamental frequency F0. The recording session included sets of words and sentences in Polish, recorded in silence, as well as in the presence of interfering signals, i.e. pink noise and so-called babble speech, also referred to as the "cocktail-party" effect. As a part of the research, speech samples were processed - both sentences and words spoken by men. The study shows that raising the fundamental frequency results in increased values of the speech quality index, measured using the PESQ (Perceptual Evaluation of Speech Quality) standard.

2

The Use of Deep Learning in Speech Enhancement

Ram R., Mohanty M. N.

Annals of Computer Science and Information Systems

|

2018

|

Vol. 14

107--111

EN

Deep learning is an emerging area in current scenario. Mostly, Convolutional Neural Network (CNN) and Deep Belief Network (DBN) are used as the model in deep learning. It is termed as Deep Neural Network (DNN). The use of DNN is widely spread in many applications, exclusively for detection and classification purpose. In this paper, authors have used the same network for signal enhancement purpose. Speech is considered for the input signal with noise. The model of DNN is used with two layers. It has been compared with the ADALINE model to prove its efficacy.

3

Impact of the duration of speech sequences on speech quality

Pocta P., Vaculík M.

Journal of Telecommunications and Information Technology

|

2007

|

nr 4

72-76

EN

This paper describes simulations of speech sequences transmission for intrusive measurement of voice transmission quality of service (VTQoS) in the environment of IP networks. The aim of the simulations was to investigate the impact of the different durations of speech sequences on speech quality from the jitter rate and packet loss point of view in IP networks. The ITU-T G.729 and ITU-T G.723.1 encoding schemes were used for the purpose of the simulations. The assessment of speech quality was realized by means of perceptual evaluation of speech quality (PESQ) algorithm. A comparison of the impact of different durations of speech sequences on speech quality and determination of the optimal duration of speech sequence for measurements of speech quality in telecommunication networks, is the aim of this paper.