The SPARC audio chunk file contains 5, possible 6 chunks:
The type of this chunk is:
The data in this chunk is simply a human readable string that describes the information found in the media independent header and the SPARC audio header chunks. The data in this chunk is not actually used by any object and is included only so that humans can be provided with a description of the data found in the other chunks (i.e., media independent and SPARC au header chunks). This chunk is not required to exist.
The type of this chunk is:
The data in the chunk is:
The media filed specifies a general type of media. For au files, this is:
The format field specifies a specific format. For au files, this is:
The device field specifies the original recording device. This field isn't really used by anything yet, but is set to one of the following:
The version field specifies what version of the cm toolkit is being used. This is currently set to:
The num_frames field contains how many frames of data are in the file.
The length_s field indicates how long the clip is in seconds.
The length_ns field indicates how many nanoseconds in addition to the previously indicated number of seconds the clip contains. Thus, the length of the clipfile can be found as (length_s + (length_ns * ONE_BILLION)).
The type of this chunk is:
The data in this chunk is:
The dataFormat field is set to one of the standard SPARC audio format codes.
The samplingRate is set to the sampling rate (samples per second).
The channelCount is set to the number of channels of audio described by the data.
The type of this chunk is:
The data of this chunk consistes of (number_of_frames + 1) entries of the following form:
The start time of the nth frame is the nth entry of the above form. Again the actual time is found as:
The end time of the nth frame is the start time of the n+1 frame. Thus, the Frame Time Table contains one more time entry than number of frames in order to specify the end time of the last frame.
The type of this chunk is:
The data of this chunk consists of number_of_frames entries of the following form:
Offset refers to the byte offset from the beginning of the data segment where the frame's data begins.
Length refers to the length of the frame's data.
Each frame's offset and length are specified requiring number_of_frames entries of this form.
The type of this chunck is:
The data of this chunk is the audio data itself.
The frame offset table is specified relative to the beginning of this chunk.