new HomeSeer install, and first impressions


Active Member
I finally got some time over the weekend and installed HomeSeer's 30 day trial software on my PC as well as a speaker client on a tablet PC I got over the holidays. This, plus the setup of two Audreys (controlled by mcsAudrey plug-in, thanks to M.McS) is my first step towards a "smart" home.

First, I must say, HomeSeer was much easier to use than MisterHouse. It has mature UPB support (which MH still can't claim, I don't think), easy event handling, and things seemed pretty straightforward. The biggest issues I had were getting my Audreys flashed correctly.

I do have a couple of questions:
1. Is there a way to get Audreys to do voice recognition similar to the HomeSeer speaker client? I'm guessing the PC running the speaker client does the speech processing, then communicates with the HomeSeer server, and that the Audreys are not powerful enough to process the speech like that. But HomeSeer seems to easily send wav files down to the Audreys, so wouldn't it be possible for the Audreys to record the voice (it does have a mic) and send the wav file up to the HomeSeer PC for processing? I realize this would have a little lag time, but wouldn't it be better than nothing?

2. Is there a way to improve/train the voice recognition from the speaker clients? It seems to misunderstand me on a regular basis.

3. Would I need additional PC's (tablets or otherwise) to run additional speaker clients? Or is there such thing as wireless mic/speaker combo (or IP mic) that I could connect to my current speaker client for input (and feedback) in another room?

I wrote down more questions somewhere else, so when I find them, I'll post them to this thread as well.
oh, I remember another question. I saw on HomeSeer that there was a place to specify customized webpages. Are these used for Audreys? Can someone point me at an example Audrey page so I would know how to make on specifically for that screen size? Do I have to point at a different URL when I want to use this page?
Homeseer speech recognition just uses Microsoft speech recognition. To train it, use the Windows Control Panel, click on Speech, and then click Train Profile.
ano said:
Homeseer speech recognition just uses Microsoft speech recognition. To train it, use the Windows Control Panel, click on Speech, and then click Train Profile.
is this done on the speaker client or on the HomeSeer server?

it would seem to be a pain if this needed to be done for each speaker client.

1 - I'm not sure if the Audry's can do speech recognition "back to" HomeSeer but Michael McSharry is the resident Audry specialist on the HomeSeer bulletin board. He's developed an Audry plugin that allows complete control of the Audry from HomeSeer. You can read about the plugins capabilities here:

There doesn't seem to be a lot of guys using Audry's anymore but this plugin was popular in it's hay day.

2 - As was already mentioned you train the speaker apps at the location that the app is installed. The idea being you train your laptop, you wife trains her pc and the kids do the same, etc...

3 - This whole "wireless mic" idea has been covered in depth on the HomeSeer bulletin board. There's an entire forum dedicated to this concept.

4 - As far as custom web pages goes you can develop your own and use the HomeSeer web server to serve this pages.
Audrey cannot do VR. It can be used as a TTS client using either the natural voices or synthesized ones Remote VR with Homeseer was viable with xapmcsVoice a few years ago. Now with HS2's remote speaker it is easier to use it unless you already have an xAP environment. Remote VR is also possible without remote computers if you want to invest in mic switching electronics.
fortunately, the touchpad interface for HS has been updated in v2.1 so that it does recognize UPB devices... this works out well for the Audreys, so I'm pretty happy with that...

also, I found a mult-user alarm clock/wake up script that seems to work out very well, so i'm very excited about that.

this just keeps getting better and better...
Michael, how do I get the Audreys to use HomeSeer variables? For example, you can make HomeSeer speaker client say "The time is $time" and it will speak the current time. Putting this same string in AudreySpeak results in "The time is dollar time".

Another general question (for everyone, not just the honorable McS), my wife has sort of a strange name for TTS. When I type in her name (Eka, short 'e'), the TTS says (Eeka, i.e., long 'e')... how can I make it pronounce (and perhaps, recognize my voice) her name correctly?