Voice API Programming Guide — June 2005 59
Call Progress Analysis
TONE_DATA.numofseg
Specifies the number of segments for a multi-segment tone.
7.8.3 Rules for Modifying a Tone Definition on DM3 Boards
Consider the following rules and guidelines for modifying default tone definitions on DM3 boards
using the voice API library:
•You m us t i ss ue dx_querytone(), dx_deletetone( ), and dx_createtone() in this order, one
tone at a time, for each tone definition to be modified.
•Attempting to create a new tone definition before deleting the current call progress tone will
result in an EDX_TNQUERYDELETE error.
•When dx_querytone( ), dx_deletetone( ), or dx_createtone( ) is issued in asynchronous
mode and is immediately followed by another similar call prior to completion of the previous
call on the same device, the subsequent call will fail with device busy.
•Only default call progress analysis tones and SIT sequences are supported for these three
functions. For a list of these tones, see Table5, “Special Information Tone Sequences (DM3)”,
on page 54 and Table6, “Default Call Progress Analysis Tone Definitions (DM3)”, on
page 57.
•These three voice API functions are provided to manipulate the call progress analysis tone
definitions. However, not all the functionality provided by these tones is available through the
voice API. You may need to use the Global Call API to access the functionality, for example,
in the case of dial tone detection and disconnect tone detection.
•If the application deletes all the default call progress analysis tones in a particular set (where a
set is defined as busy tones, dial tones, ringback tones, fax tones, disconnect tone, and special
information tones), the set itself is deleted from the board and call progress analysis cannot be
performed successfully. Therefore, you must have at least one tone defined in each tone set in
order for call progress analysis to perform successfully.
Note: The Learn Mode API and Tone Set File (TSF) API provide a more comprehensive way to manage
call progress tones, in particular the unique call progress tones produced by PBXs, key systems,
and PSTNs. Applications can learn tone characteristics using the Learn Mode API. Information on
several different tones forms one tone set. Tone sets can be written to a tone set file using the Tone
Set File API. For more information, see the Learn Mode and Tone Set File API Software Reference.
7.8.4 Rules for Using a Single Tone Proxy for a Dual Tone
A single tone proxy (also called a twin tone) acts as a proxy for a dual tone. A single tone proxy
can be defined when you run into difficulty detecting a dual tone. This situation can arise when the
two frequencies of the dual tone are close together, are very short tones, or are even multiples of
each other. In these cases, the dual tone might be detected as a single tone. A single tone proxy can
help improve the detection of the dual tone by providing an additional tone definition.
The TONE_SEG.tn_twinmin field defines the minimum frequency of the tone and
TONE_SEG.tn_twinmax field defines the maximum frequency of the tone.