|
|
Subjective and Objective Assessment of Speech Quality
Bezdíček, Martin ; Balík, Miroslav (referee) ; Vondra, Martin (advisor)
This Bachelor's thesis is focused on subjective and objective methods for evaluation of speech quality. In the first part, the quality of transmitted speech signal and its distorting effects are given. Futher, subjective and objective methods for evaluation of speech are described. The objective methods: Signal to Noise Ratio (SNR), Segmental Signal to Noise Ratio (SSNR), the methods for evaluation in spectral area (Log Likelihood Ratio Measures and Cepstral Distance) and also Perceptual Evaluation of Speech Quality (PESQ). The next part deals with the correlation analysis which is used for the comparison of subjective and objective methods. In the last part of the bachelor thesis the results of subjective and objective methods are presented. Finally on the basis of correlation analysis with the results of subjective method, the best objective method is recommended.
|
|
|
Comparison of voice and audio codecs
Lúdik, Michal ; Sysel, Petr (referee) ; Míča, Ivan (advisor)
This thesis deals with description of human hearing, audio and speech codecs, description of objective measure of quality and practical comparison of codecs. Chapter about audio codecs consists of description of lossless codec FLAC and lossy codecs MP3 and Ogg Vorbis. In chapter about speech codecs is description of linear predictive coding and G.729 and OPUS codecs. Evaluation of quality consists of description of segmental signal-to- noise ratio and perceptual evaluation of quality – WSS and PESQ. Last chapter deals with description od practical part of this thesis, that is comparison of memory and time consumption of audio codecs and perceptual evaluation of speech codecs quality.
|
|
| |
|
| |
|
|
Speech quality objective methods
Kovařík, Jiří ; Balík, Miroslav (referee) ; Sysel, Petr (advisor)
In terms of the bachelor´s thesis was created a computer programme for the quality speech estimation. This programme was created in the MATLAB programming language using GUI programing environment. Through the use of created programme was processed group of records disturbed by various noises from the real environment. As a speech quality objective methods was chosen classical SNR, Segmental SNR, PESQ, Itakura distance, Itakura – Saito distance For this programme were created service instructions. In this work are these speech quality objective methods described furthermore the process of the human voice creation, speech modeling and a linear predictive coding.
|
|
|
Audio signal denoising using deep learning
Pacal, Tomáš ; Záviška, Pavel (referee) ; Mokrý, Ondřej (advisor)
This thesis deals with noise removal in audio signal using deep learning. The basic types of neural networks and their use in audio signal processing are described. The possibilities of implementing neural networks are tested in Matlab and Python. Subsequently, a~convolutional neural network model is proposed, according to which four different convolutional network architectures are implemented and then trained and tested on different types of noise. Based on these tests, one architecture was selected and subjected to a comparative test on a speech recording and then on a music recording, together with a noise reduction method using wavelet transform. The results are evaluated using both objective sound quality metrics and an informal listening test. The neural network achieved better results according to all the metrics used as well as in the listening test.
|
|
| |
|
| |
|
|
Speech quality objective methods
Kovařík, Jiří ; Balík, Miroslav (referee) ; Sysel, Petr (advisor)
In terms of the bachelor´s thesis was created a computer programme for the quality speech estimation. This programme was created in the MATLAB programming language using GUI programing environment. Through the use of created programme was processed group of records disturbed by various noises from the real environment. As a speech quality objective methods was chosen classical SNR, Segmental SNR, PESQ, Itakura distance, Itakura – Saito distance For this programme were created service instructions. In this work are these speech quality objective methods described furthermore the process of the human voice creation, speech modeling and a linear predictive coding.
|
|
|
Comparison of voice and audio codecs
Lúdik, Michal ; Sysel, Petr (referee) ; Míča, Ivan (advisor)
This thesis deals with description of human hearing, audio and speech codecs, description of objective measure of quality and practical comparison of codecs. Chapter about audio codecs consists of description of lossless codec FLAC and lossy codecs MP3 and Ogg Vorbis. In chapter about speech codecs is description of linear predictive coding and G.729 and OPUS codecs. Evaluation of quality consists of description of segmental signal-to- noise ratio and perceptual evaluation of quality – WSS and PESQ. Last chapter deals with description od practical part of this thesis, that is comparison of memory and time consumption of audio codecs and perceptual evaluation of speech codecs quality.
|
guest ::
