Create header file for SNDAN analysis programs.

For every soundfile to be analysed by pvan or mqan, a header file must be provided. This contains documentary information about the sound. The graphics tools in SNDAN draw on this information to create relevant annotations for the plot. The file is in binary format, so cannot be printed or edited. The header information, including further numeric infomation specific to the analysis file, is displayed whenever a file is loaded by one of the analysis programs. See also showhdr.

Running mkheader.

mkheader requires several parameters, in the form of free-form text strings. Any which are not provided in the command line are requested interactively. Typing just the program name brings up a usage message. The titles for the strings demonstrates the extent to which SNDAN is assumed to be used to analyse acoustic instruments or voices. In other cases a little ingenuity may be called for. Some elements are used to annotate graphic displays; these are indicated with an asterisk. The graphic tools do not prepend titles to these strings.

mkheader  headfile [performer [instrument [date [pitch [dynamic [vibrato [part ["comments"]]]]]]]]

Parameter strings can contain spaces, but if entered on the command line the whole string must be enclosed in double quotes, as in "Tenor Saxophone". The quotes are not needed when using the program interactively. In this mode, type <Enter> to complete each string. Internally, strings are limited to 80 characters, though for tidy graphic annotations, they should be much more concise than that.


the name of the output file. Conventionally this uses the file extension .head, but this is not mandatory.
If the name of a human performer is not appropriate here (perhaps simply unknwon, or the sound is synthetic in origin, you could type "anonymous", "sample" or perhaps the name of the synthesizer, or, of course, the name of the associated soundfile.
If there is no obvious instrument name to enter here, a short description would be useful, such as "filtered noise".
There is no fixed format for this; it can be as simple as "Tuesday", or be in some convenient format such as "01:01:2000"
This is still textual material; use the musical note if appropriate, or some descriptive phrase  if not.
The program suggests musical markings such as 'mp' or 'ff'; or you could enter an amplitude in decibels (dB).
the expected entry here is either "yes" or "no".
this is intended to indicate how much of the sound is being analysed. Used interactively, the program suggests the words "all" or "part".
As its title suggests, this is a final free-form space in which to annotate the sound. Multiple lines can be used, using <Enter> in the usual way. When all the text is written, press <CTRL-z><Enter> to finish.
[Top][SNDAN home page]