chip that can do audio over IP

Hi I'm a software engineer wondering about a hardware solution. We currently have a product that uses an ATA to convert audio to IP to send to a remote device where an ATA converts digital back to audio. We dial the ATA to tel l it where we want the audio to go. We need to allow up to 63 remote devic es to simultaneously receive one-way outgoing audio (speech and maybe backg round music).

At the same time we need to allow phone conversations with two modes. The first mode is where a master can call up to 64 remote receivers simultaneou sly who can all listen but not talk back - one way all call speech. The se cond mode is where the master can talk to at least five remote listeners (p hones) simultaneously and all five can talk back to the master but don't ne ed to hear each other. Currently we use multiple ATA devices at each locat ion to achieve this.

All locations are connected via a separate protocol so we are able to send and receive information and commands between them.

Are there any chips that can turn audio into IP that would let us do the ab ove. For the larger systems we don't mind having multiple cards each with e.g. one or two audio chips on them. I found this from googling

formatting link
are but as I'm not a hardware engineer I'm not sure if this meets the requireme nts. If this chip would do, does anyone know how much they cost and whethe r it's ok from a longevity point of view - i.e. it won't become obsolete so on.

Reply to
gp.kiwi
Loading thread data ...

Wahts is an ATA - I find 133 uses of that TLA at

formatting link

Your best bet is to talk to the chip manufacturer about its price and capability. They will obviously expect to go on forever - but you would need to look into their business a bit deeper before betting your own company on their chip.

You could do it with an FPGA - but don't be fooled into thinking your software engineering background makes you ready to do it yourself - it sounds like a reasonably big FPGA job. The good thing about this route is that FPGAs tend top stay in production for long while and your design would probably port to the next generation that comes along without too much difficulty.

MK

Reply to
Michael Kellett

tly have a product that uses an ATA to convert audio to IP to send to a rem ote device where an ATA converts digital back to audio. We dial the ATA to tell it where we want the audio to go. We need to allow up to 63 remote d evices to simultaneously receive one-way outgoing audio (speech and maybe b ackground music).

The first mode is where a master can call up to 64 remote receivers simulta neously who can all listen but not talk back - one way all call speech. Th e second mode is where the master can talk to at least five remote listener s (phones) simultaneously and all five can talk back to the master but don' t need to hear each other. Currently we use multiple ATA devices at each l ocation to achieve this.

end and receive information and commands between them.

e above. For the larger systems we don't mind having multiple cards each w ith e.g. one or two audio chips on them. I found this from googling

compare

rements. If this chip would do, does anyone know how much they cost and wh ether it's ok from a longevity point of view - i.e. it won't become obsolet e soon.

oh, sorry, analog telephone adapter - with VOIP e.g.

formatting link

Thanks. I'm looking for an off the shelf solution as much as possible. The project is on the back burner because we don't have much development time to spare so I'm trying to find the quickest way of doing it.

Reply to
gp.kiwi

I'd look at teardowns of commercial ATAs or VOIP routers, and then chase up the chip manufacturers inside. For example:

formatting link
formatting link

However most of the ICs seem to be from very niche manufacturers, and availability of their silicon and documentation might be an issue (unless you can commit to xx,000 sales). Also many of them have substantial software stacks - possibly running on an adjacent SoC running Linux - so they aren't simple drop-in parts.

Zarlink (since swallowed by Microsemi swallowed by Microchip) have a range of telephony frontends, which are often used in routers to implement VOIP features - but the SIP stack still runs on the main CPU.

formatting link

Theo

Reply to
Theo

Most of the big voip players offer similar products, group paging, music on hold, listen only conference, full 2 way conference, conference moderator listen only.

I have not looked at what a asterisk server can do in a long time but that piece of software may well have a lot of these "master" capabilities. For a simplified remote ata you might be able to do the entire thing with a uC. You can get them with the ethernet phy/mac built in. You dont need the full SIP stack.

--
Chisolm 
Republic of Texas
Reply to
Joe Chisolm

tly have a product that uses an ATA to convert audio to IP to send to a rem ote device where an ATA converts digital back to audio. We dial the ATA to tell it where we want the audio to go. We need to allow up to 63 remote d evices to simultaneously receive one-way outgoing audio (speech and maybe b ackground music).

The first mode is where a master can call up to 64 remote receivers simulta neously who can all listen but not talk back - one way all call speech. Th e second mode is where the master can talk to at least five remote listener s (phones) simultaneously and all five can talk back to the master but don' t need to hear each other. Currently we use multiple ATA devices at each l ocation to achieve this.

end and receive information and commands between them.

e above. For the larger systems we don't mind having multiple cards each w ith e.g. one or two audio chips on them. I found this from googling

compare

rements. If this chip would do, does anyone know how much they cost and wh ether it's ok from a longevity point of view - i.e. it won't become obsolet e soon.

on hold,

n only.

t piece of software may

ata you might be

t phy/mac built in.

I was looking at some of the sources for telephony products and I found mor e than one who build hardware to install in a PC with a recommendation to u se asterisk for the software to tie it all together. Looks like it is a pr etty viable product commercially.

Rick C.

Reply to
gnuarm.deletethisbit

ently have a product that uses an ATA to convert audio to IP to send to a r emote device where an ATA converts digital back to audio. We dial the ATA to tell it where we want the audio to go. We need to allow up to 63 remote devices to simultaneously receive one-way outgoing audio (speech and maybe background music).

The first mode is where a master can call up to 64 remote receivers simul taneously who can all listen but not talk back - one way all call speech. The second mode is where the master can talk to at least five remote listen ers (phones) simultaneously and all five can talk back to the master but do n't need to hear each other. Currently we use multiple ATA devices at each location to achieve this.

send and receive information and commands between them.

the above. For the larger systems we don't mind having multiple cards each with e.g. one or two audio chips on them. I found this from googling

y#compare

uirements. If this chip would do, does anyone know how much they cost and whether it's ok from a longevity point of view - i.e. it won't become obsol ete soon.

c on hold,

ten only.

hat piece of software may

e ata you might be

net phy/mac built in.

ore than one who build hardware to install in a PC with a recommendation to use asterisk for the software to tie it all together. Looks like it is a pretty viable product commercially.

Yes it looks very good functionally and commercially. I suspect its hardwa re needs might be a little higher than we would like but I will check it ou t. We also need a way of continually monitoring that it's still working.

Reply to
graeme.prentice

Send a INVITE with some special From address that all your players agree to use (you control all the software, correct?). The response from the INVITE could be a 603 decline or something like that. This tells you the network and SIP stack are working. This should work outside of asterisk.

--
Chisolm 
Republic of Texas
Reply to
Joe Chisolm

e:

urrently have a product that uses an ATA to convert audio to IP to send to a remote device where an ATA converts digital back to audio. We dial the A TA to tell it where we want the audio to go. We need to allow up to 63 rem ote devices to simultaneously receive one-way outgoing audio (speech and ma ybe background music).

es. The first mode is where a master can call up to 64 remote receivers si multaneously who can all listen but not talk back - one way all call speech . The second mode is where the master can talk to at least five remote lis teners (phones) simultaneously and all five can talk back to the master but don't need to hear each other. Currently we use multiple ATA devices at e ach location to achieve this.

to send and receive information and commands between them.

do the above. For the larger systems we don't mind having multiple cards e ach with e.g. one or two audio chips on them. I found this from googling

dway#compare

requirements. If this chip would do, does anyone know how much they cost a nd whether it's ok from a longevity point of view - i.e. it won't become ob solete soon.

usic on hold,

listen only.

t that piece of software may

mote ata you might be

hernet phy/mac built in.

d more than one who build hardware to install in a PC with a recommendation to use asterisk for the software to tie it all together. Looks like it is a pretty viable product commercially.

rdware needs might be a little higher than we would like but I will check i t out. We also need a way of continually monitoring that it's still workin g.

to use (you control all

line or something like that.

utside of asterisk.

Yes, we control all the software. Thanks.

Reply to
graeme.prentice

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.