Hi,
the quick fix for this is most probably increasing the timeouts a
little. Sounds counterintuitive, but the problem appears to be that a
key press reported after a timeout in RTP DTMF packets will not be
considered again, even if its a different key (TS differs), because
it is not in a pending state any more. a better implementation is in
the pipeline, but needs some testing still.
Index: AmDtmfDetector.h
===================================================================
--- AmDtmfDetector.h (revision 1620)
+++ AmDtmfDetector.h (working copy)
@@ -345,7 +345,7 @@
// after MAX_PACKET_WAIT packets with no RTP DTMF packets received,
// a RTP DTMF event is sent out to the aggregating detector
- static const int MAX_PACKET_WAIT = 5;
+ static const int MAX_PACKET_WAIT = 8;
/**
* Time when first packet for current event was received
*/
@@ -382,7 +382,7 @@
public AmKeyPressSink
{
private:
- static const int WAIT_TIMEOUT = 100; // miliseconds
+ static const int WAIT_TIMEOUT = 200; // miliseconds
/**
* Session this class belongs to.
*/
Regards
Stefan
Medakkar, Sachin (Sachin) wrote:
Thanks for replying so quickly.
The pin I’m entering is 3309. I’ve attached 2 trace files.
The first is for the error path in which only one 3 from the pin is
captured. The second trace file is where the digits were pressed slower
than in the error scenario such that all the digits were successfully
captured as expected.
------------------------------------------------------------------------
*From:* Raphael Coeffic [mailto:[email protected]]
*Sent:* 25 January 2010 16:07
*To:* Medakkar, Sachin (Sachin)
*Cc:* [email protected]
*Subject:* Re: [Sems] Configuring DTMF Timeout
On 25.01.2010 16:51, Medakkar, Sachin (Sachin) wrote:
I’m using the Voicebox application as it comes out-of-the-box and am
defining the pin that should be prompted for in the P-App-Param header
Do you have a trace?
-Raphael.
------------------------------------------------------------------------
size=2 width="100%" align=center tabindex=-1>
*From:* [email protected]
<mailto:[email protected]>
[mailto:[email protected]] *On Behalf Of *Raphael Coeffic
*Sent:* 25 January 2010 15:47
*To:* [email protected] <mailto:[email protected]>
*Subject:* Re: [Sems] Configuring DTMF Timeout
On 25.01.2010 16:25, Medakkar, Sachin (Sachin) wrote:
Hi –
When detecting DTMF to collect a pin to access Voicebox, I’ve noticed
that if the pin contains more than 1 consecutive number that is the same
(e.g. 3309), SEMS does not recognise the number entered if I enter the
pin relatively quickly. This behaviour if not reproduced if the numbers
are different (e.g. 3039).
Could you tell me whether this is a configurable item and where this
behaviour is defined so that I may modify it as required.
Could you please tell us hwo the DTMF are transported? If these are real
tones, I'm afraid you will have to change some constants in the code.
Please search in the list archive as there should at least a couple of
threads discussing the issue.
Cheers,
Raphael.
Kindest regards,
Sachin
_______________________________________________
Sems mailing list
[email protected] <mailto:[email protected]>
http://lists.iptel.org/mailman/listinfo/sems
------------------------------------------------------------------------
_______________________________________________
Sems mailing list
[email protected]
http://lists.iptel.org/mailman/listinfo/sems
--
Stefan Sayer
VoIP Services Consulting and Development
Warschauer Str. 24
10243 Berlin
tel:+491621366449
sip:[email protected]
email/xmpp:[email protected]
_______________________________________________
Sems mailing list
[email protected]
http://lists.iptel.org/mailman/listinfo/sems
