Removing inband DTMF
Hello, Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this. Any advice? Kind regards, T.
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only
http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky <jalsot@gmail.com> wrote: -
thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
Hello, Thanks. Yeah, I know that command, as far as I know, it is only for detection, it does not change the media. Anybody else? Regards, T. On Fri, Jan 14, 2011 at 11:23 AM, Avi Marcus <Avi@amarcus.com> wrote:
http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi
On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky <jalsot@gmail.com> wrote:
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
First of, AFAIK there's nothing in FS that can currently do it. I'm not even sure it's possible at all. You can subtract a noise from audio, but the noisy nature of an encoded signal would mean it wouldn't work well (you'd still hear something), and DTMF can't be detected until some of it will have already been sent to the other endpoint so you'd still have the start of each keypress heard. For the same reason at the end until the detector spots the dtmf has ended you'll still be subtracting from the audio which will actually generate noise. The only way to avoid that may be to introduce a delay but that's not something FS can do. -Steve On 16 January 2011 15:33, Tamas Jalsovszky <jalsot@gmail.com> wrote:
Hello,
Thanks. Yeah, I know that command, as far as I know, it is only for detection, it does not change the media. Anybody else?
Regards, T.
On Fri, Jan 14, 2011 at 11:23 AM, Avi Marcus <Avi@amarcus.com> wrote:
http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi
On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky <jalsot@gmail.com> wrote:
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
That's not true. A detector can spot the start of dtmf in less than 40ms. The audio frames can then be replaced with 2833 packets. This is how it works on media gateways. I'm sure FS could be patched to do this if it uses the spandsp dtmf detector, which can detect dtmf duration. On Jan 16, 2011, at 11:50, Steven Ayre <steveayre@gmail.com> wrote:
First of, AFAIK there's nothing in FS that can currently do it.
I'm not even sure it's possible at all. You can subtract a noise from audio, but the noisy nature of an encoded signal would mean it wouldn't work well (you'd still hear something), and DTMF can't be detected until some of it will have already been sent to the other endpoint so you'd still have the start of each keypress heard. For the same reason at the end until the detector spots the dtmf has ended you'll still be subtracting from the audio which will actually generate noise. The only way to avoid that may be to introduce a delay but that's not something FS can do.
-Steve
On 16 January 2011 15:33, Tamas Jalsovszky <jalsot@gmail.com> wrote: Hello,
Thanks. Yeah, I know that command, as far as I know, it is only for detection, it does not change the media. Anybody else?
Regards, T.
On Fri, Jan 14, 2011 at 11:23 AM, Avi Marcus <Avi@amarcus.com> wrote: http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi
On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky <jalsot@gmail.com> wrote:
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
I didn't mean a significant amount of time. 40ms with a ptime of 20ms means you've already sent a packet containing inbound dtmf. -Steve On 16 January 2011 20:54, Chris Rienzo <cmrienzo@gmail.com> wrote:
That's not true. A detector can spot the start of dtmf in less than 40ms. The audio frames can then be replaced with 2833 packets. This is how it works on media gateways. I'm sure FS could be patched to do this if it uses the spandsp dtmf detector, which can detect dtmf duration.
On Jan 16, 2011, at 11:50, Steven Ayre <steveayre@gmail.com> wrote:
First of, AFAIK there's nothing in FS that can currently do it.
I'm not even sure it's possible at all. You can subtract a noise from audio, but the noisy nature of an encoded signal would mean it wouldn't work well (you'd still hear something), and DTMF can't be detected until some of it will have already been sent to the other endpoint so you'd still have the start of each keypress heard. For the same reason at the end until the detector spots the dtmf has ended you'll still be subtracting from the audio which will actually generate noise. The only way to avoid that may be to introduce a delay but that's not something FS can do.
-Steve
On 16 January 2011 15:33, Tamas Jalsovszky < <jalsot@gmail.com> jalsot@gmail.com> wrote:
Hello,
Thanks. Yeah, I know that command, as far as I know, it is only for detection, it does not change the media. Anybody else?
Regards, T.
On Fri, Jan 14, 2011 at 11:23 AM, Avi Marcus < <Avi@amarcus.com> Avi@amarcus.com> wrote:
<http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf> http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi
On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky < <jalsot@gmail.com> jalsot@gmail.com> wrote:
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
Beat your provider to death for doing something utterly stupid. If you are only acting on 2833 it won't matter if they send inband and 2833 at the same time.. It won't matter as long as you are not staring a dtmf detector on the target system and only the system converting from inband to 2833 does so. /b On Jan 16, 2011, at 3:47 PM, Steven Ayre wrote:
I didn't mean a significant amount of time. 40ms with a ptime of 20ms means you've already sent a packet containing inbound dtmf.
-Steve
Yeah, we came to this conclusion pretty quickly. We can ignore inband dtmf in case there is FS at the end. The problem is with other vendors' other devices which are not that flexible as FS is. I don't know why Cirpack is trying to push their shit over and cannot substract inband dtmf... Regards, T. On Sun, Jan 16, 2011 at 11:16 PM, Brian West <brian@freeswitch.org> wrote:
Beat your provider to death for doing something utterly stupid. If you are only acting on 2833 it won't matter if they send inband and 2833 at the same time.. It won't matter as long as you are not staring a dtmf detector on the target system and only the system converting from inband to 2833 does so.
/b
On Jan 16, 2011, at 3:47 PM, Steven Ayre wrote:
I didn't mean a significant amount of time. 40ms with a ptime of 20ms means you've already sent a packet containing inbound dtmf.
-Steve
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
Yes, that's true. I usually see 10-30ms of dtmf followed by RFC2833 packets. On Sun, Jan 16, 2011 at 4:47 PM, Steven Ayre <steveayre@gmail.com> wrote:
I didn't mean a significant amount of time. 40ms with a ptime of 20ms means you've already sent a packet containing inbound dtmf.
-Steve
On 16 January 2011 20:54, Chris Rienzo <cmrienzo@gmail.com> wrote:
That's not true. A detector can spot the start of dtmf in less than 40ms. The audio frames can then be replaced with 2833 packets. This is how it works on media gateways. I'm sure FS could be patched to do this if it uses the spandsp dtmf detector, which can detect dtmf duration.
On Jan 16, 2011, at 11:50, Steven Ayre <steveayre@gmail.com> wrote:
First of, AFAIK there's nothing in FS that can currently do it.
I'm not even sure it's possible at all. You can subtract a noise from audio, but the noisy nature of an encoded signal would mean it wouldn't work well (you'd still hear something), and DTMF can't be detected until some of it will have already been sent to the other endpoint so you'd still have the start of each keypress heard. For the same reason at the end until the detector spots the dtmf has ended you'll still be subtracting from the audio which will actually generate noise. The only way to avoid that may be to introduce a delay but that's not something FS can do.
-Steve
On 16 January 2011 15:33, Tamas Jalsovszky < <jalsot@gmail.com> jalsot@gmail.com> wrote:
Hello,
Thanks. Yeah, I know that command, as far as I know, it is only for detection, it does not change the media. Anybody else?
Regards, T.
On Fri, Jan 14, 2011 at 11:23 AM, Avi Marcus < <Avi@amarcus.com> Avi@amarcus.com> wrote:
<http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf> http://wiki.freeswitch.org/wiki/Misc._Dialplan_Tools_start_dtmf start_dtmf does detection from the inband. I'm not sure that it removes them from the rtp, though... -Avi
On Fri, Jan 14, 2011 at 12:14 PM, Tamas Jalsovszky < <jalsot@gmail.com> jalsot@gmail.com> wrote:
Hello,
Is there a way to remove an inband DTMF signal from the RTP stream with FreeSWITCH? We have a partner with ugly Cirpack which can not remove inband DTMF when the codec is g711. The problem is, that our other party wants RFC2833 only - thus we have to remove inband dtmf. Unfortunately Cirpack is not open to do this.
Any advice?
Kind regards, T.
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list <FreeSWITCH-users@lists.freeswitch.org> FreeSWITCH-users@lists.freeswitch.org <http://lists.freeswitch.org/mailman/listinfo/freeswitch-users> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:<http://lists.freeswitch.org/mailman/options/freeswitch-users> http://lists.freeswitch.org/mailman/options/freeswitch-users <http://www.freeswitch.org>http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
_______________________________________________ FreeSWITCH-users mailing list FreeSWITCH-users@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-users UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users http://www.freeswitch.org
participants (6)
-
Avi Marcus
-
Brian West
-
Chris Rienzo
-
Christopher Rienzo
-
Steven Ayre
-
Tamas Jalsovszky