Paul or Kyle, i am in the last stages of trying to get paging to work thru my cav6.6. trying to use output from my PC sound card as TTS input to the paging connection on the cav6.6. the cav6.6 recognizes the when the TTS signal is sent, but all the zones just cycle on/off quickly resulting in garbled page. can u elaborate a bit how u have successfully set up autosensing paging to the cav6.6?
thanks.
I'm wondering this myself as I haven't gotten to this point yet but I'd like to begin testing it.
Hopefully Paul will have an answer.
Kyle
There isn't a simple answer to this question because the solution is different depending on which systems are involved.
The Russound CAV6.6 takes about a half second to activate the audio paging. This delay is required because zones that are turned off (but configured to accept paging) need a short amount of time to be powered on. The result is that the first word in an announcement may not be heard.
If you’re using a security/automation controller like the Elk M1 or OmniPro, you can simply add a short tone at the beginning of each announcement to kick-start the CAV’s paging circuit (think of the tone you hear on an airplane before an announcement). If you get the duration correct, you won’t hear the tone… just the announcement.
If you’re using a software based automation package with TTS capabilities (HomeSeer, Cinemar MLSpeak, etc.), you can use a WAV tone as well or simply insert something random at the start of each line of speech.
Example: Speaking “The time is 10:07am” through the CAV’s paging input will likely result in you only hearing “time is 10:07am”.
If you instead have TTS speak “Hi, The time is 10:07am”, you’ll likely hear “The time is 10:07am” like you wanted. If by chance the paging is already on, this still sounds natural compared to starting the sentence with something arbitrary like “dog” which would accomplish the same thing but not be very appropriate if you did hear it.
To take this to the next level, the CAV6.6 paging input will also turn off quickly (~.7sec). This doesn’t cause too much of an issue with simple announcements from the Elk or Omni, but when trying to duplicate pauses in natural speech with TTS it can really make a mess. A good example would be having TTS speak weather or news events which may comprise of several sentences or more. The CAV’s paging input will often turn off at the end of a sentence and then take another .5sec to turn back on… the result can be very unsatisfactory!
My work-around for this is to use 12v to trigger the paging input instead of the audio sensing. Automation controllers like the Elk & Omni can close a relay when they are speaking using automation rules which can in turn feed 12v (using a power supply) to the CAV6.6 paging input. For example with the Elk M1 you can create an automation rule that says “Whenever audio amp turns on, then turn output X on” and then “Whenever audio amp turns off, then turn output X off”.
For systems using TTS you can do something similar… For instance with HomeSeer, they have a plug-in called “IsSpeaking” that will allow events to be triggered when TTS is speaking or idle. Again, you’d need some type of relay output (M1, Omni, ADI Ocelot w/I/O, Global Cache, etc.) that HomeSeer can control to trigger a 12v signal to the CAV’s paging.
One last bit of advice if you’re planning to use audio sensing only (no 12v trigger). The paging input has a sensitivity adjustment marked “gain” on the back of the CAV6.6. Try adjusting this for best results.
Bottom line is that this will take some trial end error to get working with most setups. Hopefully the information I've provided can help you get started in the right direction?
Cheers,
Paul