Routine

Purpose

Description

This routine reads the audio file header and optionally prints the header information. Several file header formats are supported. For files with no header or an unrecognized header, the file format can be declared by calling routine AFsetNHpar. A banner identifying the audio file and its parameters is printed.

AU audio file:
8-bit mu-law, 8-bit A-law, 8-bit integer, 16-bit integer, 24-bit integer, 32-bit integer, 32-bit IEEE floating-point, and 64-bit IEEE floating-point data formats are supported. WAVE file:
8-bit mu-law, 8-bit A-law, offset-binary 1-bit to 8-bit integer, 9-bit to 32-bit integer, 32-bit IEEE floating-point, and 64-bit IEEE floating-point data formats are supported. AIFF or AIFF-C sound file:
8-bit mu-law, 8-bit A-law, 1-bit to 32-bit integer, 32-bit IEEE floating-point, and 64-bit IEEE floating-point data formats are supported. NIST SPHERE audio file:
8-bit mu-law and 16-bit integer data formats are supported. IRCAM soundfile:
8-bit mu-law, 8-bit A-law, 8-bit integer, 16-bit integer, 32-bit integer, and 32-bit floating-point data formats are supported. ESPS sampled data feature file:
16-bit integer, 32-bit integer, 32-bit floating-point, and 64-bit floating-point data formats are supported. SPPACK sampled data file:
8-bit mu-law, 8-bit A-law and 16-bit integer data formats are supported. INRS-Telecommunications audio file:
16-bit integer format is supported. Cadence SPW Signal file:
8-bit integer, 16-bit integer, 32-bit floating-point, 64-point floating-point, and text formats are supported. CSL NSP file:
16-bit integer format is supported. Text audio file:
Data in character format (usually representing 16-bit integer values). Headerless audio file:
The data format is specified by calling routine AFsetNHpar.

For the fixed point file data representations, read operations return data values as follows. The scaling factor shown below is applied to the data in the file to give an output in the range [-1, +1).

   data format     file data values              scaling factor
  8-bit mu-law    [-32,124, +32,124]              1/32768
  8-bit A-law     [-32,256, +32,256]              1/32768
  8-bit integer   [-128, +127]                    1/256
  16-bit integer  [-32,768, +32,767]              1/32768
  24-bit integer  [-8,388,608, +8,388,607]        1/8388608
  32-bit integer  [-2,147,483,648, 2,147,483,647] 1/2147483648

File type determination:
The default behaviour is to determine the file type from the header of the input file. However this requires look-ahead and is not possible with files that are not random access (for instance a file stream from a pipe). For both random access and sequential access files, the file type can be specified explicitly with a call to routine AFsetFileType, obviating the need to determine the file type from the header.

Number of samples:
In some types of audio file types, the number of samples is specified in the file header. In others, the number of samples is known only if the file size can be determined, specifically if the file random access. For input from a stream which does not allow random access (for instance a file stream from a pipe), the file size cannot be determined for those files without that information in the file header.

                  no. samples
 file type         in header
 AU                  yes
 WAVE                yes
 AIFF or AIFF-C      yes
 NIST SPHERE         yes
 IRCAM               no
 ESPS                yes
 SPPACK              yes
 INRS-Telecom        no
 Cadence SPW         no
 CSL NSP             yes
 Text file           no
 Headerless          no

On encountering an error, the default behaviour is to print an error message and halt execution.

Parameters

Author / revision

See Also