RE: [Asterisk-Dev] Voicemail sound level threads?
On Sun, 2004-11-28 at 15:39 -0800, Kris Boutilier wrote:
> Can't say - but I do notice a significant difference between just the linear
> (wav) and gsm compressed (WAV) files you provided... I just checked a
> voicemail off of my system (CVS-v1-0-11/03/04) and I notice I'm getting the
> same effect, though I'd not realised it before. Although my attenuation
> seems less severe than yours you can really visualise the drop in energy
> levels by putting the different files through Sonogram
> (http://www.dfki.de/~clauer/programming.html)
As I keep mentioning to people, in format_wac.c, asterisk writes data
with a bit shift. From wav_write.c around line 479,
/* Volume adjust here to accomodate */
for (x=0;xdatalen/2;x++) {
tmpf = ((float)tmpi[x]) * ((float)(1 << GAIN));
if (tmpf > 32767.0)
tmpf = 32767.0;
if (tmpf < -32768.0)
tmpf = -32768.0;
tmp[x] = tmpf;
tmp[x] &= ~((1 << GAIN) - 1);
#if __BYTE_ORDER == __BIG_ENDIAN
tmp[x] = (tmp[x] << 8) | ((tmp[x] & 0xff00) >> 8);
#endif
}
I have at line 64 in the same file that GAIN = 2.
You will also see in wav_read that if you have no data in the lower bits
that asterisk assumes it had pushed the data up and will shift it back
down.
--
Steven Critchfield <[EMAIL PROTECTED]>
___
Asterisk-Dev mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
It's the default:
wav49|gsm|wav
Steve brought up the IP issue...what about .ogg? I am thinking, although
this really doesn't address the issue of the audio levels, of a format
that is not so "dead" as MS-GSM (yeah, it's kinda easy enough to pack
GSM into MS-GSM, but the only mention of it ever within the IETF was the
MS I-D). From what I recall, WMP will play .ogg (but I have not verified
this).
But I digress
Greg
p.s. this is from the -r v1-0 branch of CVS.
Kris Boutilier wrote:
Can't say - but I do notice a significant difference between just the linear
(wav) and gsm compressed (WAV) files you provided... I just checked a
voicemail off of my system (CVS-v1-0-11/03/04) and I notice I'm getting the
same effect, though I'd not realised it before. Although my attenuation
seems less severe than yours you can really visualise the drop in energy
levels by putting the different files through Sonogram
(http://www.dfki.de/~clauer/programming.html)
Can you publish the 'format=' line from /etc/asterisk/voicemail.conf off of
your server? Simply placing 'wav' as the first format in the list may be
enough to work around the problem for now - though it will cause attachment
size issues for people getting their VMs via email.
Thanks.
-Original Message-
From: Gregory Junker [mailto:[EMAIL PROTECTED]
Sent: November 28, 2004 3:25 PM
To: Asterisk Developers Mailing List
Subject: Re: [Asterisk-Dev] Voicemail sound level threads?
Just out of curiosity...was the GSM codec in Asterisk ever verified
against the ETSI reference inputs/outputs?
Greg
Gregory Junker wrote:
Any chance you could record a sample problem voicemail and
post all three
recorded formats (raw, wav and wav49) for download somewhere?
http://acd.dayark.com/downloads/msg0005.gsm
http://acd.dayark.com/downloads/msg0005.wav
http://acd.dayark.com/downloads/msg0005.WAV
{clip}
___
Asterisk-Dev mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
___
Asterisk-Dev mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
etc. I'm a non-programmer and really don't have a clue how to get closer to the problem at this point. Any thoughts? Actually, yes I do have thoughts ;) which is why I wanted to find out what had been tried/tested yet. My hypothesis involves transcoding issues between G7xx and GSM, which is why I asked in one recent post if the GSM codec had been verified against the ETSI test files, and in another, I added that the test VMs I put up were done over (I am assuming) G711u. Of course, if that has already been looked into and ruled out then I would investigate elsewhere. Greg ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
RE: [Asterisk-Dev] Voicemail sound level threads?
Can't say - but I do notice a significant difference between just the linear
(wav) and gsm compressed (WAV) files you provided... I just checked a
voicemail off of my system (CVS-v1-0-11/03/04) and I notice I'm getting the
same effect, though I'd not realised it before. Although my attenuation
seems less severe than yours you can really visualise the drop in energy
levels by putting the different files through Sonogram
(http://www.dfki.de/~clauer/programming.html)
Can you publish the 'format=' line from /etc/asterisk/voicemail.conf off of
your server? Simply placing 'wav' as the first format in the list may be
enough to work around the problem for now - though it will cause attachment
size issues for people getting their VMs via email.
Thanks.
> -Original Message-
> From: Gregory Junker [mailto:[EMAIL PROTECTED]
> Sent: November 28, 2004 3:25 PM
> To: Asterisk Developers Mailing List
> Subject: Re: [Asterisk-Dev] Voicemail sound level threads?
>
>
> Just out of curiosity...was the GSM codec in Asterisk ever verified
> against the ETSI reference inputs/outputs?
>
> Greg
>
> Gregory Junker wrote:
> >> Any chance you could record a sample problem voicemail and
> post all three
> >> recorded formats (raw, wav and wav49) for download somewhere?
> >
> >
> > http://acd.dayark.com/downloads/msg0005.gsm
> > http://acd.dayark.com/downloads/msg0005.wav
> > http://acd.dayark.com/downloads/msg0005.WAV
> >
{clip}
___
Asterisk-Dev mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
Just out of curiosity...was the GSM codec in Asterisk ever verified against the ETSI reference inputs/outputs? Greg Gregory Junker wrote: Any chance you could record a sample problem voicemail and post all three recorded formats (raw, wav and wav49) for download somewhere? http://acd.dayark.com/downloads/msg0005.gsm http://acd.dayark.com/downloads/msg0005.wav http://acd.dayark.com/downloads/msg0005.WAV ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
FWIW, the SPA3K through which this test call went, is using G711u (or at least is set to prefer it). Greg Gregory Junker wrote: Any chance you could record a sample problem voicemail and post all three recorded formats (raw, wav and wav49) for download somewhere? http://acd.dayark.com/downloads/msg0005.gsm http://acd.dayark.com/downloads/msg0005.wav http://acd.dayark.com/downloads/msg0005.WAV ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
Any chance you could record a sample problem voicemail and post all three recorded formats (raw, wav and wav49) for download somewhere? http://acd.dayark.com/downloads/msg0005.gsm http://acd.dayark.com/downloads/msg0005.wav http://acd.dayark.com/downloads/msg0005.WAV ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
RE: [Asterisk-Dev] Voicemail sound level threads?
> -Original Message-
> From: Rich Adamson [mailto:[EMAIL PROTECTED]
> Sent: November 28, 2004 1:22 PM
> To: Asterisk Developers Mailing List
> Subject: Re: [Asterisk-Dev] Voicemail sound level threads?
>
{clip}
> As sort of a short summary, I tested both the x100p and
> tdm04b cards with multiple pstn analog lines. Both exhibit the low
voicemail
> volume problem. SIP or IAX users have not incurred the same problem.
>
> In addition, Frank has opened a case with digium support
> based on the same low voicemail problem using a digium TE405 T1 card. He
has
> been working with John Bigelow (support) to document/diagnose the problem,
however
> Frank's system is not Internet exposed. As of last week, I volunteered
> to contact John and allow him to use our * system to diagnose
> the problem since I can make our system Internet accessible and both of
> us have the same problem.
{clip}
Have either of you experienced poor or sporadic performance of the zaptel
software echo canceller during interactive calls passing through the problem
interfaces? If not, how are you handling echo cancellation?
And just to clarify: you have the same problem on calls arriving on all your
available PSTN lines and all lines are terminated through the same serving
Central Office?
Any chance you could record a sample problem voicemail and post all three
recorded formats (raw, wav and wav49) for download somewhere?
Thanks.
Kris Boutilier
Information Systems Coordinator
Sunshine Coast Regional District
___
Asterisk-Dev mailing list
[EMAIL PROTECTED]
http://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] why not disable clock when using multiple Zaptel cards?
On November 28, 2004 12:28 pm, Peter Svensson wrote: > One problem that has been / is being discussed in another mail thread is > that the cards are in fact not synchronized. After a while the internal > clocks on the cards will diverge a large enough fraction of a ms. If I > remember correctly there is only on ms worth of buffering where at least > two would be needed to handle all possible phase shifts between the > interrupts on two cards. > > Even if the two (or more) cards were made to drive their clocks at exactly > the same frequency they can still end up in different phases. Of course, a > more advanced clock servo loop can keep the 1kHz timer phase. This, I > feel, is where things get tricky. Hmm -- So I suppose there is no mechanism to say "start collecting TDM data *now*" and then let the clock go at the whatever rate it's going at on the local card, picking it up whenever the interrupt occurs on the "master" clock card -- okay there will be some skew but I wonder how noticeable it really will be. This is, of course, all just generic hand-waving at this stage. :-) Similarly I'm curious as to what it would sound like if the cards did all run free and you just collected the 8 bytes of data when the master card was picking up its data -- you'd have some initial skew but after the first data 'pull' it shouldn't matter... they're all crystal-driven so the actual discrepancy between cards in the same case (i.e. at the same ambient and with the same power) would be on the order of a few (hundred maybe?) ppm -- hardly anything you'd notice in real life. -A. ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] Voicemail sound level threads?
> Are there any threads anyone can dig up (and bump) on the > voicemail-too-quiet issue? And yes, Steve, I DTFG and didn't find > anything on it, and could not find anything immediately obvious in the > archives. Greg, There have been several on asterisk-users since about July 12th when bugs 2022 and 2023 were submitted. However, since the Subject line did not necessarily include those keywords, finding them via google is almost impossible. Before opening those two bugs, I had conducted several tests. Today, I summarized those tests at: http://www.routers.com/asteriskprob/asterisk-config.htm As sort of a short summary, I tested both the x100p and tdm04b cards with multiple pstn analog lines. Both exhibit the low voicemail volume problem. SIP or IAX users have not incurred the same problem. In addition, Frank has opened a case with digium support based on the same low voicemail problem using a digium TE405 T1 card. He has been working with John Bigelow (support) to document/diagnose the problem, however Frank's system is not Internet exposed. As of last week, I volunteered to contact John and allow him to use our * system to diagnose the problem since I can make our system Internet accessible and both of us have the same problem. I've not heard back from John as yet (Thanksgiving?). In an email/irc discussion with Mark relative to the problem, he asked if I would document the test config so that a digium employee (specifically James) could "lab" it and recreate the problem. I did that on August 27th and have never heard from anyone again. Around August 29th, Mark and Russell Bryant added the app_test routines to help diagnose the problem. Mark's comment of "I can definitely measure a surprising mismatch between FXS -> FXO and FXO -> FXS paths..." in 2023 indicates an issue was observed (but not yet diagnosed/fixed) involving a T1 card, but that app apparently is not usable by those of us with digium fxo (non-T1) cards. There has been some suggestions from bkw and others this could be a hardware/motherboard issue. I don't have a clue whether it is or isn't, but I don't think anything has formally or informally been ruled out at this time. Could it be something involving dropping a significant bit, word alignment, or some other weird thing? Don't know, but it certainly suggests there are some unknown x100p/tdm motherboard/system dependencies impacting these digium cards that are unknown at this time. (For those that haven't bothered to read bug 2022 and 2023, we've all tried the tip/ring reversal, rx/tx gain adjustments, different vm recording formats, no shared interrupts, zttest = 99.987793%, no GUI, latest cvs head several times since July 2004, etc. I believe I can say our * systems have been stable otherwise. I have not tried the "raw" vm recording format as I didn't know it even existed.) I can make our * system available to anyone with the skills to diagnose the problem, including three pstn numbers to access the system, sip acct, etc. I'm a non-programmer and really don't have a clue how to get closer to the problem at this point. Any thoughts? Rich Adamson ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
[Asterisk-Dev] Voicemail sound level threads?
Are there any threads anyone can dig up (and bump) on the voicemail-too-quiet issue? And yes, Steve, I DTFG and didn't find anything on it, and could not find anything immediately obvious in the archives. Thanks Greg ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] MP1327 trunked radio channel...
Axel, I am one of the developers for app_rpt and the PCI quad radio card. You may reach me at 619-660-7838 in the evenings and weekends or 858-495-4123 during business days. We don't like putting our email addresses on the web bacause of spam. Steve Rodgers On Saturday 13 November 2004 08:57, Axel Schlueter wrote: > Hi, > > I'd like to implement a channel to connect asterisk to a > MPT1327 analogue trunked radio. I hadn't implemented something > for asterisk before, could you give me some pointers for > documentation ? I studied the implemantion of chan_oss.c, > but I'm not quite sure if it would be better to create a > new channel or instead write a zaptel device driver. Any > hints ? And one last question: I will need a "Push to talk" > functionality. Is there any infrastructure in asterisk which > supports PTT-like operation for radios ? I don't want to use a > VOX PTT. I've seen the "key/unkey radio" control frames in > the IAX2 protocol, could these be used by an IAX client to > trigger PTT ? > > TIA, > Axel ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] why not disable clock when using multiple Zaptel cards?
On Sun, 28 Nov 2004, Andrew Kohlsmith wrote: > On November 28, 2004 04:23 am, Peter Svensson wrote: > > If there are several digium cards with 1kHz interrupts then maybe only one > > of them should be causing interrupts. The other cards could be serviced > > from within that handler. It does sound a bit tricky. > > This is what I'm not getting; why it's tricky. > > Say you have an ISR with a list of cards to service. Normally the list has > only one entry: the entry for its own card. > > Now with two cards installed, the module init code would see the second card, > disable the timer interrupt on the second and add card 2's address to the > list of cards to service. One problem that has been / is being discussed in another mail thread is that the cards are in fact not synchronized. After a while the internal clocks on the cards will diverge a large enough fraction of a ms. If I remember correctly there is only on ms worth of buffering where at least two would be needed to handle all possible phase shifts between the interrupts on two cards. Even if the two (or more) cards were made to drive their clocks at exactly the same frequency they can still end up in different phases. Of course, a more advanced clock servo loop can keep the 1kHz timer phase. This, I feel, is where things get tricky. Peter ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] why not disable clock when using multiple Zaptel cards?
On November 28, 2004 04:23 am, Peter Svensson wrote:
> If there are several digium cards with 1kHz interrupts then maybe only one
> of them should be causing interrupts. The other cards could be serviced
> from within that handler. It does sound a bit tricky.
This is what I'm not getting; why it's tricky.
Say you have an ISR with a list of cards to service. Normally the list has
only one entry: the entry for its own card.
Now with two cards installed, the module init code would see the second card,
disable the timer interrupt on the second and add card 2's address to the
list of cards to service.
Something like (oversimplified I am sure):
Existing way:
init_card(void *card)
{
enable_timer(card);
}
process_interrupt(void *card)
{
process(card);
}
new way:
void *cardlist[MAX_CARDS];
static int n_cards = 0;
init_card(void *card)
{
cardlist[n_cards++] = card;
if(n_cards == 1)
disable_timer(card)
else
enable_timer(card);
}
process_interrupt(void **cardlist)
{
int i;
void *card;
for(i=0; ihttp://lists.digium.com/mailman/listinfo/asterisk-dev
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] need some suggestions
Mike M wrote: On Sun, Nov 28, 2004 at 03:24:42AM -0600, Rehan Ahmed wrote: Dear Michael, pardon my question, but do u know any one who has implimented ss7 on asterisk yet, at a reasonable cost. I do not know of any "SS7 on Asterisk" implementations. If SS7 on * is truly desired, then purchasing a card solution with embedded MTP2 is the way to go. The cost will be several thousand dollars per card. We have SS7 running with *, using Digium's cards. If you contact me, I can provide more information. Regards, Steve ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
Re: [Asterisk-Dev] need some suggestions
On Sun, Nov 28, 2004 at 03:24:42AM -0600, Rehan Ahmed wrote: > Dear Michael, > > pardon my question, but do u know any one who has implimented ss7 on > asterisk yet, at a reasonable cost. I do not know of any "SS7 on Asterisk" implementations. If SS7 on * is truly desired, then purchasing a card solution with embedded MTP2 is the way to go. The cost will be several thousand dollars per card. Achieving path redundancy will be...not worth the effort. Sharing the SS7 bandwidth with other * without redundancy increases the negative impact on users when an SS7 failure does occur. A 64 kb/s SS7 link runnig at typical max load of 40% utilization will support 1200 calls per minute. If the * needs a reset, then 1200 or less customers will be without service during the reset. The SS7 network provider expects operational stability from its connected links. If the SS7 function is on a separate signal gateway that connects to * over IP, then the s/g can serve n X * and be redundant. This approach saves on capital costs and operation costs, and raises reliability and stability to accepted levels. I do know of reasonably priced s/g that want to serve Asterisk but Asterisk is not ready to communicate with an s/g. That's why I suggest the project below. The next step after B is to break the D channel and insert two s/g (shown in C). The network application is shown in D. A-D shows the conversion of PRI FAS to PRI FAS over IP. Step E is to develop PRI NFAS over IP. Step F is to develop native ISUP over IP. Somewhere along the path SCCP/TCAP based applications are integrated (free call, billing number services, calling name delivery, special routing services, SMS, etc.). C) phone---*B channels---[ * ]---phone | | | +--D--[s/g]--SS7--[s/g]--D--+ D) phone---*IMT---pstn | | +--D---[s/g]---SS7 network > > I am setting up a new telco and need SS7 interface to itnerconnect > with the local provider > > Rehan > > > > On Sat, Nov 27, 2004 at 09:46:21AM +0500, Xenith Panacea wrote: > > > I need some advice/suggestions for as to what should we work on. > > > Something that is practical and beneficial not only for our degree > > > work, but more importanlty which contributes to the * development. > > > > Asterisk needs an IMT/SS7 interface. I am advocating an approach > > where a signal gateway interfaces the SS7 network and provides Q.931 > > signals (packets) over IP to Asterisk. For * to take advantage of > > this capability, it needs to expose its PRI FAS function to an IP > > port. > > > > Shown below, A is now. B is after development. Having B would > > advance development significantly toward having SS7 based call setup > > on Asterisk. > > > > Test: > > > > A) phone---*PRI*---phone > > > > B) phone---*B channels*---phone > > | | > > +D channel-+ > > -- > > Michael Mueller > > ___ > > Asterisk-Dev mailing list > > [EMAIL PROTECTED] > > http://lists.digium.com/mailman/listinfo/asterisk-dev > > To UNSUBSCRIBE or update options visit: > >http://lists.digium.com/mailman/listinfo/asterisk-dev > > -- Michael Mueller ___ Asterisk-Dev mailing list [EMAIL PROTECTED] http://lists.digium.com/mailman/listinfo/asterisk-dev To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-dev
