Search

The accuracy of existing auditory saliency calculation models was analyzed in this paper. The subjective auditory attention saliency of 20 pieces of sound materials was judged by psychoacoustic experiment. Then the objective saliency of these sound materials was calculated using the existing auditory saliency calculation models. According to the relationship between the calculation results of the saliency calculation models and the subjective auditory attention perception of the human ear, the accuracy of existing auditory saliency calculation models was analyzed. The experimental results showed that, although auditory saliency computational models could calculate and judge the attention saliency of sound on certain extent, they had own shortcomings when calculating different types of sounds. Choosing different computational models based on the characteristics of the sound could improve the prediction accuracy of the auditory saliency calculation models.

The previous research on visual-auditory interaction has been shown that the quality of visual stimuli may have effect on auditory perception. This paper investigated the quantitative relation between the just noticeable difference (JND) of pure tone loudness and the mean opinion score (MOS) of video quality of different bit rate. In the psychoacoustic experiments loudness JND was measured, in which visual stimuli with three kinds of bit rate were presented to subjects with audio stimuli simultaneously. MOS of video quality was evaluated by subjective assessment. The experiment results indicated that with the decrease of the video bit rate, video quality MOS declined, and loudness JND increased. From experiment results, it is concluded that loudness perception showed downtrend when MOS of video quality declined.

Classification of musical instrument is resulted from musicology. Thus, automatic identification of instrument families not only benefits the study in musicology, but also worth of attention in MIR. Based on a database consists of 2177 clips from Chinese and Western instrumental music, the experiments provided in this paper evaluate the ability of automatic Chinese and Western instrument identification by confusion matrices. Regarding various features and classifiers, the experiment results revealed the similarity and difference between Chinese and western instrumental music, which produce new possibilities for the study in identifying and classifying Chinese and Western instrument families.

Automatic discrimination of music mood is one important field of MIR. Considering the difference of Chinese traditional music and Western classical music, it is necessary to study these two kinds of music's mood taxonomy. In this paper, the mood taxonomy models of Chinese traditional music and Western classical music are implemented, and then three content feature sets are extracted directly from the waveform audio clips. Finally, music clips are classified by three feature sets and their combination using Bayesian network classifier. The experiment results indicate that the detection rate of Chinese traditional music mood taxonomy is lower than that of Western classical music mood taxonomy no matter using single feature set or their combinations, that is to say the contribution of different feature set to music mood taxonomy is different, and the detection rate improves obviously when combining three feature sets both for Chinese traditional music and Western classical music.

Auditory masking effect would be affected while visual stimulation was presented simultaneously. And quality as an important character of video should be considered. Two group simultaneous visual-auditory experiments were designed in this paper, in order to find out how visual influence auditory masking effect. The first group aimed to select video content which might influence auditory masking experiment obviously. And the second group set two different quality levels for the video content chose in the first group. Then the differences of auditory masking effects in these two groups were observed. And affective mechanism of video quality on auditory masking experiment was discussed at the end of paper.

According to numbers of music cognitive experiments, moods or emotions in music could be categorical. Since mood classifications are commonly used to structure the large collections of music available on the Web, automatic discrimination between mood taxonomy of Chinese traditional music and Western classical music would be a valuable addition to music information retrieval (MIR) systems. In this paper, three content feature sets are extracted directly from the waveform audio clips, and then two mood taxonomy models are implemented. A Bayesian network is trained to classify the discrete mood categories. Finally, because the already-known algorithms have rarely applied to the Chinese traditional music, the comparative experimental result between Chinese and Western music evokes further research necessities.

Mandarin is tonal language. For Mandarin, tone identification is very important for speech recognition and pronunciation evaluation. This paper proposes a new approach for tone identification of isolated Mandarin syllables. The system employs a voiced/unvoiced (V/UV) classifier, an improved cepstral pitch detector and a powerful tone identification method. At the same time, the system takes several efficient measures to reduce the negative effect of creaky voice. Finally, the experimental results based on a very large database show that the tone recognition system classifies four tones very well.

The tonal features can demonstrate some musical acoustical characteristics of a musical instrument. The Constant-Q Transform (CQT) transforms temporal signal into logarithmically spaced frequency domain, which suits musical content very well. To analyze the sound of the notes played on Chinese plucked string instruments, this paper proposed an algorithm to draw the chromagram used for extracting tonal features and recorded four Chinese plucked string instruments. The classic brute force method for CQT was employed to produce the spectrograms of notes which were tuned and mapped to obtain the chromagrams. Tonal features were then extracted from them and proved to be informative for analyzing the timbre of Chinese plucked string instruments.