Academia.eduAcademia.edu

Figure 22 – uploaded by Casady Bowman

See full PDFdownloadDownload figure
Figure 22. Amount of category and emotion rating data explained for each principal component This figure illustrates the category (solid) and emotion (dashed) changes in the percent explained from the first principal component, to the second principal component. This value of percent explained tells how much of the category or emotion rating data respectively, is explained by the principal component.

Figure 22 Amount of category and emotion rating data explained for each principal component This figure illustrates the category (solid) and emotion (dashed) changes in the percent explained from the first principal component, to the second principal component. This value of percent explained tells how much of the category or emotion rating data respectively, is explained by the principal component.

Related Figures (25)

Figure 1. Attack time of a waveform audio file. This figure gives an example of the acoustic component attack time, for a waveform audio file (wav). Sections a through i in the figure indicates separate attack times; this is the time in seconds from the vertical black line, to the peak of the sound indicated by the vertical red line. acoustic component attack time, for a waveform audio file (wav). Sections a through 7 in
Figure 1. Attack time of a waveform audio file. This figure gives an example of the acoustic component attack time, for a waveform audio file (wav). Sections a through i in the figure indicates separate attack times; this is the time in seconds from the vertical black line, to the peak of the sound indicated by the vertical red line. acoustic component attack time, for a waveform audio file (wav). Sections a through 7 in
Figure 2. Attack slope of a waveform audio file. This figure gives an illustration of the acoustic component attack slope. The red arrow indicates the duration (attack time) for which the attack slope is calculated.
Figure 2. Attack slope of a waveform audio file. This figure gives an illustration of the acoustic component attack slope. The red arrow indicates the duration (attack time) for which the attack slope is calculated.
Figure 3. Roll off of a waveform audio file. This figure shows the acoustic component roll off, the red segment indicates the cutoff point of 85% for the amount of high frequencies in the signal.
Figure 3. Roll off of a waveform audio file. This figure shows the acoustic component roll off, the red segment indicates the cutoff point of 85% for the amount of high frequencies in the signal.
Figure 4. Brightness of a waveform audio file. This figure shows the acoustic component brightness. To the right of the red line is the amount of energy above 1500 Hz, or the brightness of the sound.
Figure 4. Brightness of a waveform audio file. This figure shows the acoustic component brightness. To the right of the red line is the amount of energy above 1500 Hz, or the brightness of the sound.
Figure 5. Mel-frequency cepstral coefficients (mfcc) of a waveform audio file. This figure shows the acoustic component mfcc. Each bar represents the numerical (rank coefficient) value computed for the thirteen components returned.
Figure 5. Mel-frequency cepstral coefficients (mfcc) of a waveform audio file. This figure shows the acoustic component mfcc. Each bar represents the numerical (rank coefficient) value computed for the thirteen components returned.
Figure 6. Instrument judgment experiment example. Participants rated each sound on all 10 instruments.
Figure 6. Instrument judgment experiment example. Participants rated each sound on all 10 instruments.
Figure 7. Emotion judgment experiment example. Participants rated each sound on all five emotions.
Figure 7. Emotion judgment experiment example. Participants rated each sound on all five emotions.
Figure 8. Principal component analysis of instrument ratings. This figure illustrates the method used for PCA to reduce the dimensions of instrument ratings. Figure 8 A shows the original data while Figure 8 B demonstrates the reduction of the original data into principal components. The actual size of original data in Figure 8 part A and B have been decreased by the number of sounds for purposes of explanation. components that account for a certain amount of variance within the data.
Figure 8. Principal component analysis of instrument ratings. This figure illustrates the method used for PCA to reduce the dimensions of instrument ratings. Figure 8 A shows the original data while Figure 8 B demonstrates the reduction of the original data into principal components. The actual size of original data in Figure 8 part A and B have been decreased by the number of sounds for purposes of explanation. components that account for a certain amount of variance within the data.
Figure 9. Scree plot of observations for principal components describing instrument ratings. This figure demonstrates the variance in timbre ratings for each principal component of Experiment la. Percent variance accounted for by each principal component is indicated by a point on the red line. The blue line indicates cumulative percent variance for the principal components. The first two principal components account for more than 80% of the variance in the instrument rating data.
Figure 9. Scree plot of observations for principal components describing instrument ratings. This figure demonstrates the variance in timbre ratings for each principal component of Experiment la. Percent variance accounted for by each principal component is indicated by a point on the red line. The blue line indicates cumulative percent variance for the principal components. The first two principal components account for more than 80% of the variance in the instrument rating data.
Figure 10. Scree plot of observations for principal components of emotion ratings. This figure shows the variance in emotion ratings for each principal component for Experiment 1b. Percent variance explained for each principal component is specified by a point on the red line, while cumulative percent variance is indicated by the blue line. The first two principal components account for more than 80% of the variance in the emotion rating data.
Figure 10. Scree plot of observations for principal components of emotion ratings. This figure shows the variance in emotion ratings for each principal component for Experiment 1b. Percent variance explained for each principal component is specified by a point on the red line, while cumulative percent variance is indicated by the blue line. The first two principal components account for more than 80% of the variance in the emotion rating data.
Figure 11. Scree plot of observations for principal components describing [ADS category ratings. This figure shows the variance in JADS category ratings for Experiment 2a. The percent variance accounted for by each principal component is noted by a point on the red line, the blue line shows the cumulative percent variance for by the principal components. The first three principal components account for more than 80% of the variance in the IADS category rating data. Experiment 2a. The percent variance accounted for by each principal component is noted
Figure 11. Scree plot of observations for principal components describing [ADS category ratings. This figure shows the variance in JADS category ratings for Experiment 2a. The percent variance accounted for by each principal component is noted by a point on the red line, the blue line shows the cumulative percent variance for by the principal components. The first three principal components account for more than 80% of the variance in the IADS category rating data. Experiment 2a. The percent variance accounted for by each principal component is noted
Figure 12. Scree plot of observations for principal components describing [ADS emotior ratings. This figure shows the variance in IADS emotion ratings for each principal component of Experiment 2b. The percent variance accounted for by each principal component is indicated by a point on the red line. The blue line shows the cumulative percent variance accounted for by the principal components. The first three principal components account for more than 80% of the variance in the emotion rating data.
Figure 12. Scree plot of observations for principal components describing [ADS emotior ratings. This figure shows the variance in IADS emotion ratings for each principal component of Experiment 2b. The percent variance accounted for by each principal component is indicated by a point on the red line. The blue line shows the cumulative percent variance accounted for by the principal components. The first three principal components account for more than 80% of the variance in the emotion rating data.
Figure 13. Box plot of observations for timbre ratings. This figure illustrates the timbre ratings for the Experiment la. Each box indicates one instrument rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 13. Box plot of observations for timbre ratings. This figure illustrates the timbre ratings for the Experiment la. Each box indicates one instrument rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 14. R-squared. Instrument principal component two. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the proportion of R-squared for each additional predictor to the model. Figure 14a demonstrates these values of R-squared for principal component one, and Figure 14b shows the values of R- squared for principal component two for the instrument rating data. that was contributed for each addition of a predictor to the model for instrument ratings.
Figure 14. R-squared. Instrument principal component two. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the proportion of R-squared for each additional predictor to the model. Figure 14a demonstrates these values of R-squared for principal component one, and Figure 14b shows the values of R- squared for principal component two for the instrument rating data. that was contributed for each addition of a predictor to the model for instrument ratings.
Significant acoustic components for instrument PCA
Significant acoustic components for instrument PCA
Figure 15. Box plot of observations for emotion ratings. This figure illustrates emotion ratings for Experiment 1b. Each box indicates one emotion rated by participants, the median is indicated by the red line, and the edges show the 25" and 75" percentiles. Whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 15. Box plot of observations for emotion ratings. This figure illustrates emotion ratings for Experiment 1b. Each box indicates one emotion rated by participants, the median is indicated by the red line, and the edges show the 25" and 75" percentiles. Whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 16 shows the proportion of R-squared contributed for each addition of a
Figure 16 shows the proportion of R-squared contributed for each addition of a
Figure 16. R-squared. Emotion principal component two. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the proportion of R-squared for each additional predictor to the model. Figure 16a demonstrates these values of R-squared for principal component one, and Figure 16b shows the values of R- squared for principal component two for the emotion rating data.
Figure 16. R-squared. Emotion principal component two. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the proportion of R-squared for each additional predictor to the model. Figure 16a demonstrates these values of R-squared for principal component one, and Figure 16b shows the values of R- squared for principal component two for the emotion rating data.
Figure 17 displays the porportion of R-squared for each of the principal components for both the instrument and emotion. This figure represents the percent of
Figure 17 displays the porportion of R-squared for each of the principal components for both the instrument and emotion. This figure represents the percent of
Figure 17. Amount of instrument and emotion rating data explained for each principal component. This figure illustrates the instrument (solid) and emotion (dashed) changes in the percent explained from the first principal component, to the second principal component. This value indicates how much of the instrument data, or emotion rating data, is explained by the principal component. and emotion principal component two decreases considerably.
Figure 17. Amount of instrument and emotion rating data explained for each principal component. This figure illustrates the instrument (solid) and emotion (dashed) changes in the percent explained from the first principal component, to the second principal component. This value indicates how much of the instrument data, or emotion rating data, is explained by the principal component. and emotion principal component two decreases considerably.
Figure 18. IADS judgment experiment example. Participants rated each sound for all 7 categories. strongly disagree, and 7 strongly agree, see Figure 18.
Figure 18. IADS judgment experiment example. Participants rated each sound for all 7 categories. strongly disagree, and 7 strongly agree, see Figure 18.
Figure 19. Box plot of observations for IADS category ratings. Each box indicates one category rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 19. Box plot of observations for IADS category ratings. Each box indicates one category rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 20. Box plot of observations for IADS emotion ratings. Each box indicates one emotion rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 20. Box plot of observations for IADS emotion ratings. Each box indicates one emotion rated by participants, the median is indicated by the red line in the center of each box, and the edges indicate the 25" and 75" percentiles. The whiskers of each plot indicate the extreme data points, and outliers are plotted outside of the whiskers.
Figure 21. R-squared. Emotion principal component three. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the value of R- squared for each additional predictor to the model. No component could explain emotion for principal component one. Figure 23a demonstrates change in amount of R-squared in the model for principal component two for the emotion rating data. Figure 23b demonstrates change in amount of R-squared in the model for principal component three for the emotion rating data.
Figure 21. R-squared. Emotion principal component three. This figure illustrates the change in R-squared for each addition of a predictor to the model. The dashed line in each figure demonstrates cumulative change, and the solid line represents the value of R- squared for each additional predictor to the model. No component could explain emotion for principal component one. Figure 23a demonstrates change in amount of R-squared in the model for principal component two for the emotion rating data. Figure 23b demonstrates change in amount of R-squared in the model for principal component three for the emotion rating data.
Related topics:
EmotionMel-frequency Cepstral Coefficients
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved