BLU-HIF Sidetone AEC dialtone
kevinzz
Posts: 35
Hello,
I have done about ten BLU-101 with BLU-HIF setups in various conference and training rooms. Overall I think it's a really slick solution.
One thing that I'm not quite keen on has to do with the behaviour of the AEC when a dialtone is active.
As soon as the phone goes off-hook and a dialtone is heard over the speakers, any sounds picked up by the mics causes the dialtone to get choppy/garbley.
I have a mixer crosspoint knob that controls the level of the mic mix (or speaker mix) that is sent to the HIF AEC reference (for the headset sidetone). I try to find at point at which the AEC functions adequately (the mic mix no longer is heard over the speakers) but below the point where the dialtone choppiness starts. I can't seem to find it.
Once the dialtone goes away, the AEC works great and speech sounds fine. I know that no one is actually conferencing while the dialtone is on, but it leaves an odd impression on the client when I have to say to just ignore it, and their sacred Polycom/Cisco/whatever phone does not do the same thing.
Any suggestions? (aside from modifying the AEC algorithm, which I think will not happen) I thought about placing a ducker between the mic mix to telephone Tx, and the trigger would be a feed from the Rx that was filtered to exactly the dialtone frequency. Crude, and would take some trial and error to get the frequncy and attack just right, but it might work.
Thanks for any insight.
Kevin
I have done about ten BLU-101 with BLU-HIF setups in various conference and training rooms. Overall I think it's a really slick solution.
One thing that I'm not quite keen on has to do with the behaviour of the AEC when a dialtone is active.
As soon as the phone goes off-hook and a dialtone is heard over the speakers, any sounds picked up by the mics causes the dialtone to get choppy/garbley.
I have a mixer crosspoint knob that controls the level of the mic mix (or speaker mix) that is sent to the HIF AEC reference (for the headset sidetone). I try to find at point at which the AEC functions adequately (the mic mix no longer is heard over the speakers) but below the point where the dialtone choppiness starts. I can't seem to find it.
Once the dialtone goes away, the AEC works great and speech sounds fine. I know that no one is actually conferencing while the dialtone is on, but it leaves an odd impression on the client when I have to say to just ignore it, and their sacred Polycom/Cisco/whatever phone does not do the same thing.
Any suggestions? (aside from modifying the AEC algorithm, which I think will not happen) I thought about placing a ducker between the mic mix to telephone Tx, and the trigger would be a feed from the Rx that was filtered to exactly the dialtone frequency. Crude, and would take some trial and error to get the frequncy and attack just right, but it might work.
Thanks for any insight.
Kevin
0
Comments
AEC struggles with Dial Tone because the signal is constant sine waves, and because some of this signal is reflected back through the telephone interface. The AEC is trying to remove some of this signal.