Speaker clients + Russound

[Cmiller]

Is there a way to send TTS to both speaker clients and Russound speakers? Does BLSpeech support adding hs4 speaker clients as BLSpeech speakers?
I have Android touch clients, Sonos Clients, and Russound (MCA88).
The Russound integration works, but leaves the standard clients silent.

Thank you!
---Chris

[Eman]

Android clients as with HStouch I guess but for proper Android integration you need the Tasker Plug-in : More on that here : https://forums.homeseer.com/forum/ul...o-speech-event


Eman.

[Cmiller]

Thank you Eman - I should have been more clear.

I have the HS touch speakers working, and the sonos speakers working.

The issue is that as soon as I register the speaker proxy in BLSpeech, only russound speaks.

In the config options->Speaking->"Speaker Client to use as main speaking client" - I get to pick only one. If I leave it blank I expected it to speak on all of them, but that doesn't seem to work. If I set it to the speaker client connected to the russound input, it works great - but only through Russound. If I set it to another speaker(Sonos, TS, etc.) - it will speak out of only that speaker. I just can't get it to speak out of all of them...

Thank you!
---Chris

[Eman]

Try naming them in settings (for testing). If they still don't work then that could be a HomeSeer issue....





I think the naming format should be as in the image below....??? Depends on how you named your devices (some people use the post code feature in HSTouch for their naming)









Eman.

[Cmiller]

In the HS events - the 'leave unchecked' works great. This works:



the 'default' one is the one connected to Russound's input.

The problem is that the BLSpeech proxy intercepts all speech and redirects it.

If I turn off the proxy, then HS TTS speaks to all these HS speakers above.

If I turn it on, it only speaks to the single one I selected in BLSpeech.



If I turn the BLSpeech proxy off, then Russound doesn't change all the output zones to the source of the "..DEFAULT" HS client and after speaking change them back. Basically, so if music is playing, it gets interrupted for a TTS event - then the music turned back on where it left off.

I feel like I need either:

1.) this BLSpeech option for 'speaker client to use as main speaking client' to accept (like HS events) "leave unchecked for all"

or

2.) the ability to specify these HS Touch devices as BLSpeech speakers.

there is an option to set an HS device as a speaker, but I am not sure how/if an HS device is/can be a speaker client. "Other?"



or... maybe I am missing something

Thank you!

---Chris

[Eman]

In the HS events - the 'leave unchecked' works great. This works:



the 'default' one is the one connected to Russound's input.

The problem is that the BLSpeech proxy intercepts all speech and redirects it.

If I turn off the proxy, then HS TTS speaks to all these HS speakers above.

If I turn it on, it only speaks to the single one I selected in BLSpeech.



If I turn the BLSpeech proxy off, then Russound doesn't change all the output zones to the source of the "..DEFAULT" HS client and after speaking change them back. Basically, so if music is playing, it gets interrupted for a TTS event - then the music turned back on where it left off.

I feel like I need either:

1.) this BLSpeech option for 'speaker client to use as main speaking client' to accept (like HS events) "leave unchecked for all"

or

2.) the ability to specify these HS Touch devices as BLSpeech speakers.

there is an option to set an HS device as a speaker, but I am not sure how/if an HS device is/can be a speaker client. "Other?"



or... maybe I am missing something

Thank you!

---Chris

The best person to debug (troubleshoot) this is randy if Blade is not picking up on this... Or else, the BLSpeech assumes you are using ceiling speakers so zoning is routed through its proxy....

But not to confuse you, send a PM either upstatemike (his handy work! https://forums.homeseer.com/forum/ho...ultiroom-audio) or the gentlemen above.


Eman.

[Uncle Michael]

If you want to increase the chance that Bob will see your question you can try the help desk on his web site. There's a link in his signature.
For reference, it is: http://www.bladeplugins.com/helpdesk/Login.asp

[upstatemike]

But not to confuse you, send a PM either upstatemike (his handy work! https://forums.homeseer.com/forum/ho...ultiroom-audio) or the gentlemen above.


Eman.

Sorry but I am not familiar with the BLSpeech proxy and how it works. It sounds like you need to use it to get the zone control features you need with Russound so the question seems to be how to send announcements to other speaker clients besides Russound when the BLSpeech proxy is enabled. I'm sure you are not the first person who has needed to do this so hopefully there is a simple answer.

I avoid the logic issues by using hardware solutions to preempt zones for announcements but I think if you are using Russound the BLSpeech proxy is likely going to be the best option if you can find the right configuration.

[Cmiller]

Thank you!

I avoid the logic issues by using hardware solutions to preempt zones for announcements but I think if you are using Russound the BLSpeech proxy is likely going to be the best option if you can find the right configuration.

Russound has a 12V trigger and a paging input. This is what I used to use with the OMNI.. (RIP) - but I don't (currently) have a way to trigger it. Is this 'trigger' what you are doing? I know I can go down the Arduino or Pi path, but it seems like more hassle managing another OS and device for something this simple. Numato has a super simple USB->GPIO solution - but no HS integration that I can find.

[upstatemike]

Thank you!



Russound has a 12V trigger and a paging input. This is what I used to use with the OMNI.. (RIP) - but I don't (currently) have a way to trigger it. Is this 'trigger' what you are doing? I know I can go down the Arduino or Pi path, but it seems like more hassle managing another OS and device for something this simple. Numato has a super simple USB->GPIO solution - but no HS integration that I can find.

I use HTD 1240ADS amplifiers which have audio sensing inputs. There is a Voice assistant input for each zone plus an overall Alarm or HA Announcement input. I feed the zone inputs from Monoprice passive multi-room controllers (just a preamp version of the Monoprice multi-room amp). The voice assistant inputs override the multi-room zone inputs and the Announcement input overrides everything. I can use audio directly from Homeseer to the announce input, or in my case feed it from a mixer so I can share it with Homeseer, Elk Security Panel, and Stargate. There are dip switches to exclude zones from all-announce (like outside zones) and I use logic in the various platforms to control what time of day announcements can happen.

If I was using a Russound amp I would probably put an audio sensing relay on the audio line out of my Homeseer server and use that to trip the "page trigger in" on the amp. Not saying that is the best solution for you but it is the way I would probably do it.

https://www.newark.com/mcm-custom-au...tor/dp/29X0259

[dcorsus]

Cmiller

See if I can help with some aspects. From the pictures, is it correct you are running this on HS4 but use some plugings that are HS3 native, but I believe you are using the HS4 version of the Sonos(4) plugin, correct?

Probably the easiest way around is, is to do 2 speak something actions in a single event.

When you select all the clients, is outcome/issue the same as selecting none? I had a discussion a while back with Rich from HomeSeer, arguing that "not selecting anything would be equal to selecting all", that this lead to issues and confusions, there was an issue back then that got resolved subsequently, but don't recall what now the final implementation is.

The way that proxies work for HS3 plugins versus the recommendation for HS4 plugins is very different. In HS3, a plugin registers itself to HS3 as a speaker proxy. All these proxies are sorta called in a daisy chained fashion, each time a plugin gets called, the plugin itself decides whether to continue the chain or stop. So in your case, it appears that as soon as you select default (BLSpeech), the BLSpeech plugin decides that the announcement was for its plugin and terminates the "hand-down-the-chain" event.

In HS4 styled plugins, plugin authors have been recommended to register speaker clients (as opposed to proxies) and I was under the impression that there is no more a "chain" and elected speakerclients get a callback. rjh Rich, how do HS3 and HS4 speaker proxies/clients interact? What is your implementation?

Now all of us plugin authors doing speaker proxy kind of plugins under HS3, I assume all of us knew about this and we all built a setting or code to deal with speak events with multiple speaker clients. Blade how did you support this?

As I wrote above, in case the code is not there, you should be able to do multiple speak action in a single event, at least as a work around if this is not just a setting thing.

edit: found the posting on the issue https://forums.homeseer.com/forum/de...60#post1450060

[Cmiller]

Cmiller

See if I can help with some aspects. From the pictures, is it correct you are running this on HS4 but use some plugings that are HS3 native, but I believe you are using the HS4 version of the Sonos(4) plugin, correct?

Probably the easiest way around is, is to do 2 speak something actions in a single event.

p=1450060#post1450060[/url]

Yes - have an S6 running HS4 with mostly HS4 plugins include Sonos, BLRussound and BLSpeech.

I have tried two actions, but once the proxy is active it doesn't care what output speaker is defined. Actually, if I define anything, it never speaks. If I define nothing (uncheck all) it speaks through BLSpeech to the assigned BLSpeech-Speakers only.

If BLSpeech only 'proxied' TTS destined for the HS speaker it had defined as its primary speaker client - that might also work.

Thank you all for your suggestions and help.

[dcorsus]

rjh Rich I think this is an issue!

I simulated the same behavior easily using my Sonos (HS3 speakerproxy baseline) and Sonos4 (HS4 speakerclient baseline) plugins.

Sonos is based on speaker proxy and when I configure my Sonos plugin to: upon a callback via SpeakIn, I configure it to not call hs.SpeakProxy I will never see any speakerclient callbacks into my Sonos4 plugin. Shouldn't you treat registered proxies different from registered speakerclients?

So in this case reported by Cmiller as soon as a speakerproxy callback happens, which I suspect is what BLSpeech picks up, if the PI consumes the announcement and don't pass the information up the chain via hs.SpeakProxy, none of the other selected speakerclients get called, or at least that's how it looks. I do suspect there is missing functions in Bob's BLSpeech pluging, because he should check on the "host" information of the SpeakIn callback, that there are other SpeakerClients and therefore, irrespective whether it generates its own announcement actions, it should call hs.SpeakProxy so the other clients get a callback.

[upstatemike]

Did you test the behavior using two separate events? What would be the downside of doing it that way?

[dcorsus]

Did you test the behavior using two separate events? What would be the downside of doing it that way?

I realized that once BLSpeech's proxy is active, it seems to tee of any speakerclient, and in doing so, it really doesn't seem to matter what speaker clients you select in the event.
So my view: 2 issues:
1. HomeSeer should call all speaker clients that are selected, independant of what happens with all the proxying
2. BLSpeech should only pick out the events meant for this PI and forward through a call to SpeakerProxy if it isn't