Introduction to DVD maps
| < PREVIOUS: Dynamic Vocal Maps _______________________________________ | > NEXT: Developmental histograms |
We use the term DVD maps for a variety of graphical methods to view vocal events dynamically at multiple time scales. Like the single frame of video, the sonogram presents a static representation of a dynamic process. The sonogram can only capture short term changes in sound (over time scales of milliseconds), whereas song development is a process occurring over time scales of vocal change ranging from minutes to
weeks. The study of that process requires the ability (a) to store large amounts of vocal data, (b) to analyze that data, and (c) to display the results of that analysis as visual representations that highlight key features of the development process as they occur over time. The availability of digital recording and the low cost of digital data storage have now made it possible to meet these requirements. The feature batch module of SAP2011 creates syllable tables including the entire vocal ontogeny of a bird on file. The DVD maps will allow you to see those data for tracing vocal changes over time.
We will use a syllable table of song development of a bird called 109. Please start by downloading it here: http://soundanalysispro.com/download/download-syllable-table-example. Then open SAP2011 and select the 'DVD maps' option. Click 'Open syllable or raw features table' and select the table syll_R109:
Note that the syll_ phrase indicates that this is a syllable table, where each record (line) summarize features of one syllable (duration, mean pitch, etc). Now click start, and you should see a movie of scatter plots of the syllable features distribution in 2D, keep looking how it changes over time.
The online analysis of vocalization and the automatic generation of the syllable table is the corner stone of SAP2011, and one of the most useful descriptive model of such tables is displaying dynamic vocal development (DVD) maps, showing vocal changes as movies. You should keep in mind, however, that segmentation of sound to syllable units is a double-edge sword: it can uncover song structure if properly used, but it might also inappropriately chop the sound into inappropriate units. This is not only because the criteria for segmenting sounds are not always robust, but mainly because the deep structure of vocalization bout is not trivially related to syllable units. The binary feature files that are automatically created during the live analysis keep record of non segmented (continuous) feature curves. SAP2011 does not provide a means of explicitly analyzing these data, but you can easily export them to Matlab.