Re: [linux-audio-user] asterisk open source linux pbx

I've recently been conscripted into the role of backup to our primary phone system admin with an eye towards leveraging my network admin background in our coming VoIP deployment. Due to this I've been cramming as much telephony information as I can into my brain over the past 3-4 weeks. Our PBX (private branch exchange, i.e. the switch for our office phone system for those unfamiliar with the term) is a proprietary system from NEC. Having to deal with systems like this is contrary to my nature. To sooth the irritation of being rubbed the wrong way by this proprietary technology I've signed on to the asterisk users and dev lists with a few aims. First, as I expected, I've found the members of the asterisk community to really know their stuff when it comes to the standards and protocols that make phone systems work. In a few days of reading the lists I've already learned much that I haven't gotten out of the NEC manuals, but that helps me to better understand their closed system. Second, I'm hoping that down the road I can get an asterisk system into this operation to provide additional services and functionality that would be more expensive to purchase from NEC.

I'm writing to LAU to get some feedback on a number of possiblities that come to mind for cross fertilization between this community and the asterisk community. Also I'm hoping there are some here who have experiences with asterisk they would be willing to share.

As I understand it asterisk can use ALSA supported full-duplex cards to provide voice i/o. An asterisk server with a number of connections to the phone network and several RME HDSP or other such high channel count multi-channel cards would seem to be a very useful, cost effective and high quality solution for supporting call-in shows and telephone interviews for a radio station. Such a settup could also provide a nice platform for an intercom system for a business or even a home. Is anyone here doing such things?

Apropos my recent inquiry regarding bats, telephony has traditionally saved bandwidth by limiting the frequencies transmitted to a roughly 4kHz band since the information important to intelligible speech can be conveyed without the sounds outside that band. Are the concepts used to capture bandlimitted audio for speech the same, or similar to, what would be used to capture the interesting information from sounds produced by animals who hear above the human hearing range? There are a variety of audio data compression and synthesis/resynthesis schemes in use in the telephony world. Have any of these been repurposed for use as effects, perhaps wrapped up in LADSPA plugins? Are there similarities between jack and asterisk in what they need to do to provide audio routing and scheduling? Perhaps this has already occured or perhaps their needs are too different, but could the two projects benefit from sharing ideas or even code? If these are naive questions and the two domains are orthogonal I'm interested in knowing why. Hmm ... as I write I realize a big difference is that many phone conversations happen at once and have no need for synchronization. jack's typical application space involves keeping many channels of audio in sync. So, I guess I've largely answered this one for myself. But, still input from the system programming gurus like Paul and Jack would be most welcome and surely enlightening and educational.

I had a few other ideas and questions, but they've slipped away from me. Anyway, this has gotten long enough. Thanks in advance for reading and for any feedback. -Eric Rz.