Re: [Freeswitch-dev] VAD, TALK and NOTALK events
The algorithm used is not so great and could stand to be replaced by a better one if we get the chance. basically we decode the audio to signed linear 16bit samples for each packet we receive and iterate the samples and add up the absolute values then compute the average and compare it to the acceptable level considered "talking" then we use a counter to make sure it's in the same state several times in a row before actually changing the status of the channel to talking or not talking. A real vad algorithm would also be able to filter out the voice from the background noise and judge the energy based only on that. Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: KMTDBOURSE <kmtdbourse@free.fr> To: freeswitch-dev@lists.freeswitch.org Sent: Thursday, February 14, 2008 1:51:55 PM Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events Hi, I ma interested in the VAD algorithm used where where is the code exactly? maybe a link any document support about the vad alogorithm used therein? thanks, Zelda At 01:47 14/02/2008, Anthony Minessale wrote: They are enabled by the vad params in the sofia profile xml <param name="vad"> Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: Cesar Cepeda <cesar@auronix.com> To: freeswitch-dev@lists.freeswitch.org Sent: Wednesday, February 13, 2008 10:17:23 AM Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events Please correct me if I’m wrong. The events are indeed fired, but only if the SWITCH_VAD_FLAG_EVENTS_TALK or NOTALK flags are set for an rtp_session. I found no place in the code where those flags are set. That is what I meant by not being “enabled”. I subscribed to the events and received nothing. Then I set the flags on switch_rtp_enable_vad function, and then I started receiving the events. Perhaps there is another way of doing it that I’m missing. Cesar Cepeda. De: freeswitch-dev-bounces@lists.freeswitch.org [ mailto:freeswitch-dev-bounces@lists.freeswitch.org] En nombre de Anthony Minessale Enviado el: Miércoles, 13 de Febrero de 2008 08:17 Para: freeswitch-dev@lists.freeswitch.org Asunto: Re: [Freeswitch-dev] VAD, TALK and NOTALK events The events are fired. I don't know what you mean by "enabled" You can either bind an event to a callback in the load function of your custom module. or connect to mod_event_socket on tcp and request "events talk notalk" void my_event_handler(switch_event_t *event) { } if (switch_event_bind(modname, SWITCH_EVENT_TALK, SWITCH_EVENT_SUBCLASS_ANY, my_event_handler, NULL) != SWITCH_STATUS_SUCCESS) { switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_ERROR, "Couldn't bind!\n"); return SWITCH_STATUS_GENERR; } if (switch_event_bind(modname, SWITCH_EVENT_NOTALK, SWITCH_EVENT_SUBCLASS_ANY, my_event_handler, NULL) != SWITCH_STATUS_SUCCESS) { switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_ERROR, "Couldn't bind!\n"); return SWITCH_STATUS_GENERR; } Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: Cesar Cepeda <cesar@auronix.com> To: freeswitch-dev@lists.freeswitch.org Sent: Tuesday, February 12, 2008 6:53:03 PM Subject: [Freeswitch-dev] VAD, TALK and NOTALK events Hi,
From looking at the code of switch_rtp.c I can see that VAD is done if
enabled in the sip profile in order to not send packets when there is no noise. There are also two events {TALK, NOTALK} which can be fired when VAD kicks in, nevertheless these two events are enabled nowhere (or at least I didn't find a place in the code where they were enabled), so no matter if you ask for them you don't receive them. My questions are the following: Is there some particular reason for not enabling these two events? Is the event queue easily overflowed if these two events are enabled? Thanks. Cesar Cepeda. _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org Never miss a thing. Make Yahoo your homepage. -----Inline Attachment Follows----- _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org Looking for last minute shopping deals? Find them fast with Yahoo! Search. _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org -----Inline Attachment Follows----- _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org ____________________________________________________________________________________ Looking for last minute shopping deals? Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping
Anthony, That begs the question - do we have VAD (or AMD or FMD) on other endpoints? Specifically I'd be interested in knowing about OpenZAP. Can it or is there any kind of machine detection or human detection (like Asterisk's AMD app) for OZ or any other endpoints capable of dialing an outbound call? Thanks, MC ________________________________ From: freeswitch-dev-bounces@lists.freeswitch.org [mailto:freeswitch-dev-bounces@lists.freeswitch.org] On Behalf Of Anthony Minessale Sent: Friday, February 15, 2008 6:09 AM To: freeswitch-dev@lists.freeswitch.org Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events The algorithm used is not so great and could stand to be replaced by a better one if we get the chance. basically we decode the audio to signed linear 16bit samples for each packet we receive and iterate the samples and add up the absolute values then compute the average and compare it to the acceptable level considered "talking" then we use a counter to make sure it's in the same state several times in a row before actually changing the status of the channel to talking or not talking. A real vad algorithm would also be able to filter out the voice from the background noise and judge the energy based only on that. Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: KMTDBOURSE <kmtdbourse@free.fr> To: freeswitch-dev@lists.freeswitch.org Sent: Thursday, February 14, 2008 1:51:55 PM Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events Hi, I ma interested in the VAD algorithm used where where is the code exactly? maybe a link any document support about the vad alogorithm used therein? thanks, Zelda At 01:47 14/02/2008, Anthony Minessale wrote: They are enabled by the vad params in the sofia profile xml <param name="vad"> Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: Cesar Cepeda <cesar@auronix.com> To: freeswitch-dev@lists.freeswitch.org Sent: Wednesday, February 13, 2008 10:17:23 AM Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events Please correct me if I'm wrong. The events are indeed fired, but only if the SWITCH_VAD_FLAG_EVENTS_TALK or NOTALK flags are set for an rtp_session. I found no place in the code where those flags are set. That is what I meant by not being "enabled". I subscribed to the events and received nothing. Then I set the flags on switch_rtp_enable_vad function, and then I started receiving the events. Perhaps there is another way of doing it that I'm missing. Cesar Cepeda. De: freeswitch-dev-bounces@lists.freeswitch.org [ mailto:freeswitch-dev-bounces@lists.freeswitch.org <mailto:freeswitch-dev-bounces@lists.freeswitch.org> ] En nombre de Anthony Minessale Enviado el: Miércoles, 13 de Febrero de 2008 08:17 Para: freeswitch-dev@lists.freeswitch.org Asunto: Re: [Freeswitch-dev] VAD, TALK and NOTALK events The events are fired. I don't know what you mean by "enabled" You can either bind an event to a callback in the load function of your custom module. or connect to mod_event_socket on tcp and request "events talk notalk" void my_event_handler(switch_event_t *event) { } if (switch_event_bind(modname, SWITCH_EVENT_TALK, SWITCH_EVENT_SUBCLASS_ANY, my_event_handler, NULL) != SWITCH_STATUS_SUCCESS) { switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_ERROR, "Couldn't bind!\n"); return SWITCH_STATUS_GENERR; } if (switch_event_bind(modname, SWITCH_EVENT_NOTALK, SWITCH_EVENT_SUBCLASS_ANY, my_event_handler, NULL) != SWITCH_STATUS_SUCCESS) { switch_log_printf(SWITCH_CHANNEL_LOG, SWITCH_LOG_ERROR, "Couldn't bind!\n"); return SWITCH_STATUS_GENERR; } Anthony Minessale II FreeSWITCH http://www.freeswitch.org/ ClueCon http://www.cluecon.com/ AIM: anthm MSN:anthony_minessale@hotmail.com GTALK/JABBER/PAYPAL:anthony.minessale@gmail.com IRC: irc.freenode.net #freeswitch FreeSWITCH Developer Conference sip:888@conference.freeswitch.org iax:guest@conference.freeswitch.org/888 googletalk:conf+888@conference.freeswitch.org pstn:213-799-1400 ----- Original Message ---- From: Cesar Cepeda <cesar@auronix.com> To: freeswitch-dev@lists.freeswitch.org Sent: Tuesday, February 12, 2008 6:53:03 PM Subject: [Freeswitch-dev] VAD, TALK and NOTALK events Hi,
From looking at the code of switch_rtp.c I can see that VAD is done if enabled in the sip profile in order to not send packets when there is no noise. There are also two events {TALK, NOTALK} which can be fired when VAD kicks in, nevertheless these two events are enabled nowhere (or at least I didn't find a place in the code where they were enabled), so no matter if you ask for them you don't receive them. My questions are the following:
Is there some particular reason for not enabling these two events? Is the event queue easily overflowed if these two events are enabled? Thanks. Cesar Cepeda. _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-dev <http://lists.freeswitch.org/mailman/options/freeswitch-dev> http://www.freeswitch.org Never miss a thing. Make Yahoo your homepage. <http://us.rd.yahoo.com/evt=51438/*http:/www.yahoo.com/r/hs> -----Inline Attachment Follows----- _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE: http://lists.freeswitch.org/mailman/options/freeswitch-dev <http://lists.freeswitch.org/mailman/options/freeswitch-dev> http://www.freeswitch.org Looking for last minute shopping deals? Find them fast with Yahoo! Search. <http://us.rd.yahoo.com/evt=51734/*http:/tools.search.yahoo.com/newsearch/category.php?category=shopping> _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org <http://www.freeswitch.org/> -----Inline Attachment Follows----- _______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org ________________________________ Never miss a thing. Make Yahoo your homepage. <http://us.rd.yahoo.com/evt=51438/*http:/www.yahoo.com/r/hs>
Hi, Thanks a lot for describing the vad algo it seems rather a basic algo and I am surprised it is not more sophisticated. I hear also from speex users (must be same algo?) that the vad algo is very poor and should be improved. That is what I am doing, try to improve vad and echo cancellation algorithms. I would appreciate a link where I can find data files (audio) to download on which I can perform tests. Of course I will feedback the freeswitch community in case of improvment. best Zelda At 15:09 15/02/2008, Anthony Minessale wrote:
The algorithm used is not so great and could stand to be replaced by a better one if we get the chance.
basically we decode the audio to signed linear 16bit samples for each packet we receive and iterate the samples and add up the absolute values then compute the average and compare it to the acceptable level considered "talking" then we use a counter to make sure it's in the same state several times in a row before actually changing the status of the channel to talking or not talking.
A real vad algorithm would also be able to filter out the voice from the background noise and judge the energy based only on that.
Hi, I had been involved on VAD few years back, now I am in the management and out of coding. But VAD still attracts me, keep me posted on the development. We had used VAD for "AIR FORCE" whenever the PILOT want to speak to Operator at nearest Control Tower. Thanks in advance Regards Rajeev Vats From: freeswitch-dev-bounces@lists.freeswitch.org [mailto:freeswitch-dev-bounces@lists.freeswitch.org] On Behalf Of KMTDBOURSE Sent: Saturday, February 16, 2008 8:18 AM vats To: freeswitch-dev@lists.freeswitch.org Subject: Re: [Freeswitch-dev] VAD, TALK and NOTALK events Hi, Thanks a lot for describing the vad algo it seems rather a basic algo and I am surprised it is not more sophisticated. I hear also from speex users (must be same algo?) that the vad algo is very poor and should be improved. That is what I am doing, try to improve vad and echo cancellation algorithms. I would appreciate a link where I can find data files (audio) to download on which I can perform tests. Of course I will feedback the freeswitch community in case of improvment. best Zelda At 15:09 15/02/2008, Anthony Minessale wrote: The algorithm used is not so great and could stand to be replaced by a better one if we get the chance. basically we decode the audio to signed linear 16bit samples for each packet we receive and iterate the samples and add up the absolute values then compute the average and compare it to the acceptable level considered "talking" then we use a counter to make sure it's in the same state several times in a row before actually changing the status of the channel to talking or not talking. A real vad algorithm would also be able to filter out the voice from the background noise and judge the energy based only on that.
KMTDBOURSE wrote:
Hi,
Thanks a lot for describing the vad algo it seems rather a basic algo and I am surprised it is not more sophisticated. I hear also from speex users (must be same algo?) that the vad algo is very poor and should be improved. That is what I am doing, try to improve vad and echo cancellation algorithms. I would appreciate a link where I can find data files (audio) to download on which I can perform tests. Of course I will feedback the freeswitch community in case of improvment.
best
Zelda There is a pretty good free echo canceller at www.rowetel.com, although FreeSwitch can't directly use it as it is GPL software. It does, however, mean that free and open workable algorithms for robust EC do exist.
VAD is different. It is still a research topic. You can see this from the way the VAD algorithms in things like G.729 have been revised over the codec's lifetime, as further research has been done. The algorithms G.729 and AMR use are encumbered by patents, otherwise cloning them would be excellent start. VAD for speech recognition is relatively straightforward. When you can buffer the audio, and look forwards and backwards in time, it become much easier to identify the start and end of voice bursts. It becomes hard when you can only look back in a low latency scenario. Steve
On Feb 18, 2008, at 6:55 PM, Steve Underwood wrote:
There is a pretty good free echo canceller at www.rowetel.com, although FreeSwitch can't directly use it as it is GPL software. It does, however, mean that free and open workable algorithms for robust EC do exist.
VAD is different. It is still a research topic. You can see this from the way the VAD algorithms in things like G.729 have been revised over the codec's lifetime, as further research has been done. The algorithms G.729 and AMR use are encumbered by patents, otherwise cloning them would be excellent start. VAD for speech recognition is relatively straightforward. When you can buffer the audio, and look forwards and backwards in time, it become much easier to identify the start and end of voice bursts. It becomes hard when you can only look back in a low latency scenario.
Steve
Isn't that Echo canceler live in Kernel space? If so then its fine. /b
Brian West wrote:
On Feb 18, 2008, at 6:55 PM, Steve Underwood wrote:
There is a pretty good free echo canceller at www.rowetel.com, although FreeSwitch can't directly use it as it is GPL software. It does, however, mean that free and open workable algorithms for robust EC do exist.
VAD is different. It is still a research topic. You can see this from the way the VAD algorithms in things like G.729 have been revised over the codec's lifetime, as further research has been done. The algorithms G.729 and AMR use are encumbered by patents, otherwise cloning them would be excellent start. VAD for speech recognition is relatively straightforward. When you can buffer the audio, and look forwards and backwards in time, it become much easier to identify the start and end of voice bursts. It becomes hard when you can only look back in a low latency scenario.
Steve
Isn't that Echo canceler live in Kernel space? If so then its fine.
That's why I said "directly" use it. I don't know what you guys intend to do about openzap drivers. Steve
Know that (OSLEC) pretty well. Wanna test new algorithms but need wav data. Echo Cancel is also still under research process, VAD is part of a good echo canceller as it freezes the algo processing in case of double-talk. Generally a modified Geigel algo or energy based algo are used. I have new algorithms and want to test them. Unfortunately, I don't have enough data, there are out there very expensive databases such as (AURORA...) but don't think they are really suitable for my tests. Zelda At 01:55 19/02/2008, you wrote:
KMTDBOURSE wrote:
Hi,
Thanks a lot for describing the vad algo it seems rather a basic algo and I am surprised it is not more sophisticated. I hear also from speex users (must be same algo?) that the vad algo is very poor and should be improved. That is what I am doing, try to improve vad and echo cancellation algorithms. I would appreciate a link where I can find data files (audio) to download on which I can perform tests. Of course I will feedback the freeswitch community in case of improvment.
best
Zelda There is a pretty good free echo canceller at www.rowetel.com, although FreeSwitch can't directly use it as it is GPL software. It does, however, mean that free and open workable algorithms for robust EC do exist.
VAD is different. It is still a research topic. You can see this from the way the VAD algorithms in things like G.729 have been revised over the codec's lifetime, as further research has been done. The algorithms G.729 and AMR use are encumbered by patents, otherwise cloning them would be excellent start. VAD for speech recognition is relatively straightforward. When you can buffer the audio, and look forwards and backwards in time, it become much easier to identify the start and end of voice bursts. It becomes hard when you can only look back in a low latency scenario.
Steve
_______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org
KMTDBOURSE wrote:
Know that
(OSLEC)
pretty well. Wanna test new algorithms but need wav data. Echo Cancel is also still under research process, VAD is part of a good echo canceller as it freezes the algo processing in case of double-talk. Generally a modified Geigel algo or energy based algo are used.
VAD and double talk are totally different problems.
I have new algorithms and want to test them. Unfortunately, I don't have enough data, there are out there very expensive databases such as (AURORA...) but don't think they are really suitable for my tests.
If you are serious about VAD testing, the TIA has a downloadable database of hand tagged audio. If you are serious about echo cancellation, you will start by getting your canceller through the G.168 tests, and then look at the real world test cases that fail. No expensive databases are needed in either case. I can't imagine a standard database even being very useful for line echo cancelller development. You will probably want to work out the difference between VAD and double talk first, though. Steve
Ya really serious about algorithms and tests. That is my job, I am doing a PhD thesis on that matter. My concern is mathematical and statistical aspects of vad and echo canceller. I don't know much about what is in technology; I have been surfing for a while but encountered a lot patented algorithms so don't know what is behind? I would appreciate documents on what is most recently implemented in gsm (cellular phones) and other technologies requiring vad and echo cancellation. had a look to http://www.vocal.com/ I see no database there, just a doc sheet about recommendation which is nor rigorous neither of any help. if u have any direct link for such databases I would appreciate it very much. best Zelda At 00:44 20/02/2008, you wrote:
KMTDBOURSE wrote:
Know that
(OSLEC)
pretty well. Wanna test new algorithms but need wav data. Echo Cancel is also still under research process, VAD is part of a good echo canceller as it freezes the algo processing in case of double-talk. Generally a modified Geigel algo or energy based algo are used.
VAD and double talk are totally different problems.
I have new algorithms and want to test them. Unfortunately, I don't have enough data, there are out there very expensive databases such as (AURORA...) but don't think they are really suitable for my tests.
If you are serious about VAD testing, the TIA has a downloadable database of hand tagged audio. If you are serious about echo cancellation, you will start by getting your canceller through the G.168 tests, and then look at the real world test cases that fail. No expensive databases are needed in either case. I can't imagine a standard database even being very useful for line echo cancelller development. You will probably want to work out the difference between VAD and double talk first, though.
Steve
_______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org
Dear Steve: Thanks a lot for your e-mail, I got database 250 and seems very helpful. I will let you know. The interesting point in my technique is that: 1) the statistical test is theoretically characterized: I derived the null na non-null hypothesis (I wonder whether this sounds to you). 2) I have the exact theoretical threshold, this is fixed theoretically not experimentally 3) I also have what we call the power of the test. These 3 points make it unique, that is what I think. It works very well in noisy environment (very noisy). I think you maybe of help to judge my results, can you? best, Zelda At 00:44 20/02/2008, you wrote:
KMTDBOURSE wrote:
Know that
(OSLEC)
pretty well. Wanna test new algorithms but need wav data. Echo Cancel is also still under research process, VAD is part of a good echo canceller as it freezes the algo processing in case of double-talk. Generally a modified Geigel algo or energy based algo are used.
VAD and double talk are totally different problems.
I have new algorithms and want to test them. Unfortunately, I don't have enough data, there are out there very expensive databases such as (AURORA...) but don't think they are really suitable for my tests.
If you are serious about VAD testing, the TIA has a downloadable database of hand tagged audio. If you are serious about echo cancellation, you will start by getting your canceller through the G.168 tests, and then look at the real world test cases that fail. No expensive databases are needed in either case. I can't imagine a standard database even being very useful for line echo cancelller development. You will probably want to work out the difference between VAD and double talk first, though.
Steve
_______________________________________________ Freeswitch-dev mailing list Freeswitch-dev@lists.freeswitch.org http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev http://www.freeswitch.org
participants (6)
-
Anthony Minessale
-
Brian West
-
KMTDBOURSE
-
Michael Collins
-
Rajeev Vats
-
Steve Underwood