www.homeseer.com    
 

Go Back   HomeSeer Message Board > HomeSeer Products & Services > HomeSeer Software > HomeSeer Voices

HomeSeer Voices Discussions related to the HomeSeer add-on voices and their use within HomeSeer.

Reply
 
Thread Tools Display Modes
  #21  
Old September 6th, 2016, 05:15 PM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
why do you write it out "speaker dot exe" instead of speaker.exe

dunno. maybe it's just my hands doing this without thinking?

First thing relating to getting speaker dot exe to run in Wine is getting SAPI to run in Wine.

I didn't have much issues with that part.

After that it's getting both TTS and VR to run in Wine.

I don't think it'll run on an RPi3. But you never know.
__________________
- Pete

Automator

HS3 Pro & Lite Edition Beta 3.0.0.435

HS3 Wintel Touch | Ubuntu 16.04 64 bit | Oracle Windows Virtual Box ==> for Wintel only SAPI and HS3 plugins | Speech - Microsoft SAPI - Neospeech - Amazon Echo | Hardware | Haswell Intel iSeries 3 - 16Gb | Pine64 - 2Gb computers | Openpeak Intel Atom SoC tabletop touchscreens (15 HS tabletop tablets) | Touchscreens - Windows embedded POE connected |Light switches - X10,UPB, ZWave and Zigbee | Firewall - PFSense - 2 WAN plus 4 LAN interfaces | Network - Gb managed switches / POE WAP(s) | CCTV - Zoneminder IPHD cams - variety | Audio - Russound - AB8SS | Security - Leviton HAI Omni Pro 2 | Weather - Davis Vantage Vue - MeteoStick - WeeWx | 1-Wire - AAG, Midon and HB | OWFS - Mosquitto - Node Red - Python - RPi Stretch - OpenWRT

Last edited by Pete; September 6th, 2016 at 05:39 PM.
Reply With Quote
  #22  
Old September 6th, 2016, 10:53 PM
RandyInLA's Avatar
RandyInLA RandyInLA is offline
Seer
 
Join Date: Aug 2016
Location: Los Angeles
Posts: 44
Quote:
Originally Posted by Pete View Post
why do you write it out "speaker dot exe" instead of speaker.exe

dunno. maybe it's just my hands doing this without thinking?
I thought it might have something to do with the forum converting it to a link. I've seen other things I've typed out become a link here and there for some reason.

The whole beauty of running the Linux version of HS on an Rpi3 was that it was a tiny, self-contained server. If I have to install SAPI & speaker.exe on another machine for voice recognition, it kinda defeats the "tiny, self-contained" part. Then I think about your post where you said you had 15+ touch pads... WOW!

I downloaded the PC version of HS onto a Windows 7 64-bit machine at work today just so I could see what the SAPI speech recognition app & speaker.exe app looked like.
Reply With Quote
  #23  
Old September 8th, 2016, 01:11 AM
RandyInLA's Avatar
RandyInLA RandyInLA is offline
Seer
 
Join Date: Aug 2016
Location: Los Angeles
Posts: 44
I was able to get speaker.exe working in a Windows XP virtual machine, as well as a Windows 8 virtual machine. Worked better in the latter, however, it was definitely too slow in both VMs. Had to speak very slowly or it didn't understand me. One cool thing was that even though the TTS was coming out of speaker.exe in the Windows VM (stuttering), the remote Ivona.com TTS I setup on the pi was still triggering out of the Pi's headphone jack, just like when I sent TTS to HS Touch. And as slow & stuttery as the Win8 VM TTS was, it still started playback of speech a full second before I heard the Ivona.com voice on the Pi. I hadn't had access to VR yet to get a good sense of the latency of sending text to ivona and waiting for a returned audio file to play. I think the novelty of using TTS remote via ivona.com in realtime has worn off after hearing the delay. Should be almost instant running locally on screaming fast hardware.

So I decided to buy an Intel NUC i7 computer w/Windows 10 Pro today to house the VR. The more I researched small footprint PCs, the more I thought about other Windows-based apps I currently run in a VM that would be much easier to use on an actual, semi-current spec'd machine... Which lead to me thinking I might as well re-install HS3 in this new Win10 box to make everything much faster, not just the VR. I'm sad to be giving up the Rpi3 HS. It has a special place in my heart, for sure. I enjoy spending most of my time in a terminal window and I don't think I'll be able to do that with Win10. The Nuc should be more than powerful enough to handle HS, complete with VR/TTS and a few other apps I may throw at it - 3.33GHz 6-core i7 w/16GB RAM & 500GB HD. Slightly better than an Rpi3 Oh! I also found the ivona voice I like, Brian, as a downloadable font for use in the Windows SAPI! Best of everything if I can get all this configured in a single box! Thanks again for all the suggestions! I'll keep the Rpi3 around to be the Homebridge server, since it's already running on it.
Reply With Quote
  #24  
Old September 8th, 2016, 06:16 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
Yeah here ran TTS/VR (but didn't use VR) on the Homeseer box of the 90's. This box was an overclocked single core AMD 400 Mhz CPU with 128/256Mb of memory. It was slower than today's RPi2 CPU and had less memory but ran AT&T SAPI TTS just fine.

I have old serially connected touchscreens here today that run off a 200 Mhz ARM CPU doing just fine.

Remember that while you can get a bunch of stuff running on an ARM based 900Mhz CPU with 1Gb of ram it is not an Intel CPU and it was never meant to run Windows applications.

Here while I am doing VM's running wintel stuff. I have one tabletop touchscreen next to the mothership running HSTouch, SAPI and VR. Using the tabletop to the zoned audio would work just fine. Two instances of speaker dot exe also work fine (from two HS3 boxes).

Relating to the Zee and Zee-2 (RPi and RPi2) here updated my RPi2 to Jessie and turned it in to a ZNet like device running X10, Z-Wave and UPB remotely to the new Zee-2 mothership.

The new mini Zee-2 mothership now is running on an Xi3/Xi5A dual core AMD 1.8Ghz CPU with 2Gb of RAM and an SSD drive for the base Ubuntu 16.04 64bit OS. It did come with Windows 7 on it.

This is a separate endeavor than the HS3 Pro mothership. I am also testing the Pine64 / 2Gb machine as it runs circles around the RPi3 and has a better features set.

IE: here never used blue tooth for any Homeseer automation. I have never ran HS on wireless and never plan to. So the adds of Bluetooth / Wireless to the RPi3 didn't do anything for me. Personally I would have liked to see an optional POE connection on the RPi3.

Last edited by Pete; September 8th, 2016 at 03:49 PM.
Reply With Quote
  #25  
Old December 19th, 2017, 01:48 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Pete

Having read through this and other threads, I am still not clear how HS3 TTS functionality integrates with MS SAPI Voices.

On my HS3 Ubuntu 16.04 box, I have installed Oracle VB and I am now preparing to install Win XP Embedded to run SAPI Voices.

What I want to do is to use HS TTS events (Speak Something) to output TTS on the HS Speaker clients using SAPI Voices.

Is this possible and if so how do you interface the two systems together? Is there a script that goes into speaker.sh or speak_to_file.sh ?
__________________
Intel NUC 7i5bnh/Kingmax 8GB/Transcend 32GB SSD | Lubuntu 16.04.3 64bit | Mono 5.8.0.108
HomeTroller SEL (Backup HomeSeer) | Lubuntu 16.04.3 64bit | Mono 5.8.0.108
HomeSeer: HS3 Standard, 3.0.0.418 | HSTouch enabled
Plugins: Z-Wave, CM15A, BLBackup, BLLAN, PHLocation, SDJ-Health, UltraLog3, EasyTrigger, Jon00VDGraphing, Jon00PluginSetV2, drhsEventEnabler, BLPlex, Chromecast, Device History
Smart Speakers: 2 X Google Home; 2 X Google Home Mini
Chromecasts: 4 X Chromecast 2
CCTV: Xeoma
Reply With Quote
  #26  
Old December 19th, 2017, 02:21 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
Speaker dot exe works using Microsoft SAPI which does TTS and VR.

For Linux Rich has incorporated Linux TTS / VR which is not SAPI.

For your Oracle VB install
1 - XPe
2 - MS SAPI 5.X - might already be there and here utilize Neospeech
3 - test your speech properties in the speech applet in control panel before running speaker.
4 - a directory in the C:\Program Files\HS3Speaker
5 - put all of your speaker files there - configure your speaker to connect to the IP of your main box and use a name / password you have configured. You will see it connect.
6 - auto start speaker dot exe on boot.

The TTS will be directed to your Ubuntu Audio. HS3 will recognize the speaker instance on your XPe VB.

Very simple. The Speaker client just connects to your HS3 running on Linux.

I have four speaker clients running in the VB.

In the network tab it looks like this when HS3 connects to the speaker application running on the VB.

Code:
Speaker	HS3W2003VB:DEFAULT
Here is what my speech control panel looks like. First one is called default and you name the next instances of speaker.

Name:  speech.jpg
Views: 45
Size:  122.8 KB

4 instances talk to two HS3 boxes.
I also have speaker running on all of the HSTouch screens (Windows) but typically do not utilize speaker rather use the Russound zoned amps.
Name:  speech2.jpg
Views: 45
Size:  12.8 KB
Name:  speakeroptions.jpg
Views: 45
Size:  106.6 KB

BTW can you please do me a favor.

Relating to HS 3 beta version 396 I saw a weird sound error when starting HS3 via the ./go command.

Can you try manually starting HS3 via a terminal session and tell me if you see this error.

Code:
ALSA lib pcm_dmix.c:1029:(snd_pcm_dmix_open) unable to open slave
aplay: main:722: audio open error: No such file or directory

Last edited by Pete; December 19th, 2017 at 02:49 AM.
Reply With Quote
  #27  
Old December 19th, 2017, 02:44 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Thanks Pete. I think I am thinking about this from a fundamentally different way which is why I am confused.

From your explanation, I understand the SAPI voice will be invoked when HS3 sends TTS to speaker.exe (which is a HS3 speaker client). So if I setup an event with 'Speak Something', I can only output SAPI voice to one or more instances of the speaker.exe (as a HS3 speaker client). Correct?

That is not the application I am thinking off. I am thinking essentially of a replacement for flite or pico2text where a TTS event using 'Say Something' outputs to multiple types of speaker clients, eg. GH speakers in my case.

Reply With Quote
  #28  
Old December 19th, 2017, 02:51 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
So if I setup an event with 'Speak Something', I can only output SAPI voice to one or more instances of the speaker.exe (as a HS3 speaker client). Correct?

yes

The speaker client that HS3 sees when booting takes over from the flite / pico2text stuff.

Here start the VB / speakers first, then a short delay then start HS3 such that on boot it connects to the VB speakers.

If you shut off the VB then the TTS will go to flite/pico2text.

So you are sending your HS3 TTS to Google Home speakers? Is the TTS in sync with multiple Google Home Speakers?

BTW here do not utilize any speaker proxy plugins. Audio goes from HS3 box to Russound zoned amps.
Reply With Quote
  #29  
Old December 19th, 2017, 03:03 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
GH Speakers = Google Home

Will I be able to output to Google Home?

Attached is a working event using 'Say Something' that sends output to multiple Google Home speakers using flite voices. This is the application that I was hoping I could use SAPI as a replacement for flite.
Attached Images
 
Reply With Quote
  #30  
Old December 19th, 2017, 03:08 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
The send to speaker commands are identical and the same as in Windows HS3.

So you shouldn't have an issue with Google Home speakers and the above event.
Reply With Quote
  #31  
Old December 19th, 2017, 03:21 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
OK, I'll give it a go. My upcoming holiday project

Thanks so much for helping with your knowledge.
Reply With Quote
  #32  
Old December 20th, 2017, 07:56 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Quote:
Originally Posted by Pete View Post

For your Oracle VB install
1 - XPe
2 - MS SAPI 5.X - might already be there and here utilize Neospeech
3 - test your speech properties in the speech applet in control panel before running speaker.
4 - a directory in the C:\Program Files\HS3Speaker
5 - put all of your speaker files there - configure your speaker to connect to the IP of your main box and use a name / password you have configured. You will see it connect.
6 - auto start speaker dot exe on boot.
OK Pete. I've got as far as Step 4 and now I am a little confused. Step 5. What speaker files are you referring to? Where do I get them from?

I have the Speech applet in Control Panel and all seems to be working fine.
Reply With Quote
  #33  
Old December 20th, 2017, 08:40 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
Way back for speaker dot exe I copied files from the Homeseer 3 directory and just created a directory called hs3. It really doesn't matter with the speaker program.

Here is what mine looks like. There was a speaker install download but I do not see it anymore.

For a manual install copy over the root bin, grammar and wave file directory.

Name:  speaker1.jpg
Views: 40
Size:  167.8 KB

Copied over the speaker installation program that you see above. It is zipped and should work for you.
Attached Files
File Type: zip SetupHS3Speaker_3_0_0_1.zip (799.6 KB, 4 views)

Last edited by Pete; December 20th, 2017 at 08:52 AM.
Reply With Quote
  #34  
Old December 20th, 2017, 08:49 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Can you share the files?

Edit: Sorry, I was on mobile and didn't see that you had attached the files. Thanks so much. Much appreciated!
Reply With Quote
  #35  
Old December 20th, 2017, 08:59 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
Attached. It's old and works fine. Never have updated it.

For multiple instances you can create separately named directories.

HSpeakerA, HSpeakerB, HSpeakerC in the

C:\Program Files\ directory.

OK so in review...

1 - on your VB test basic sounds to make sure you have an audio connect from the VB to the main Linux OS.
2 - validate SAPI going to the control speech applet on your VB. I do not test the VR stuff and just the TTS stuff
3 - Go to your VB and run speaker dot exe and connect to the IP of your HS3 box, use a user listed in your HS3 GUI.
4 - test the speech again.

I make the Wintel VB auto start on Linux reboot. I also utilize the R2 start utility for it.

On the main Linux boot of the OS I put a pause on the HS3 start up so that it sees the VB speaker running while booting HS3 in Linux.

For autobooting Linux Oracle VB, just drag the VB icon to the desktop and configure boot to autostart the VB.

Note too there base OS on HS3 Pro runs on a 160Gb SSD and base OS on HS3 Lite runs on a 32Gb SSD.

Noticed that if I start HS3 in a terminal session on the Linux box then I hear both Linux TTS and Microsoft SAPI TTS at the same time.

On autoboot of the box though I only hear the Oracle Wintel VB SAPI TTS. I am never on the console anyhow...

All should be OK.

If you are using a plugin sound proxy then all should work just fine. Here audio goes out to the Russound zoned amps. I do audio checking of my events and triggers rather than looking at the console sometimes here.

Another thing you can do with Oracle VB's is run Microsoft only plugins like the Kinect. Here run it and HSTouch on a PipoX7 connected to a 17" multitouch monitor with the Kinect sitting on top of it. A bit much though relating to the footprint.

Last edited by Pete; December 20th, 2017 at 09:30 AM.
Reply With Quote
  #36  
Old December 20th, 2017, 09:54 AM
happnatious1's Avatar
happnatious1 happnatious1 is offline
Seer Master
 
Join Date: Mar 2004
Location: Southern Maryland
Posts: 1,287
I've made some progress getting better voices on my Ubuntu install.

I downloaded Amy from http://www.cepstral.com/en/personal and ran the installer, then by changing speak.sh in my homeseer folder to:

#!/bin/sh
#flite -voice kal16 -t "$1"
#pico2wave -w=temp.wav "$1"
#aplay temp.wav
padsp /opt/swift/bin/swift -n Amy "$1"

Amy now speaks but complains that she is not licensed. I'm in contact with Cepstral to see how to remove that. It looks like they have a Linux arm version as well.
Reply With Quote
  #37  
Old December 20th, 2017, 10:11 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Quote:
Originally Posted by Pete View Post
Way back for speaker dot exe I copied files from the Homeseer 3 directory and just created a directory called hs3. It really doesn't matter with the speaker program.

Here is what mine looks like. There was a speaker install download but I do not see it anymore.

For a manual install copy over the root bin, grammar and wave file directory.

Attachment 65294

Copied over the speaker installation program that you see above. It is zipped and should work for you.
Hit a glitch. The speaker.exe is failing to initialise and crashes on launch. The install has way fewer files than indicated in your folder. Any ideas?
Attached Images
  
Reply With Quote
  #38  
Old December 20th, 2017, 10:15 AM
Pete's Avatar
Pete Pete is online now
OverSeer
 
Join Date: Jan 2001
Location: House
Posts: 15,135
Yeah better to copy the files you see above in the JPG of the speaker directory.

Originally here started with the entire Homeseer directory then deleted files not needed.

IE: just install Homeseer 3 for Windows on your VB, then delete all of the extra files in the directory while concurrently keeping what you need for running HS3.

Also make sure you have installed the dot net stuff on your VB. Thinking Dot Net 3.5 should suffice.

Thinking if you install HS3 it will install all of the dot net stuff automagically.

Soon you will be letting your fingers do the walking relating to this stuff.....

Last edited by Pete; December 20th, 2017 at 10:26 AM.
Reply With Quote
  #39  
Old December 20th, 2017, 10:17 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Quote:
Originally Posted by happnatious1 View Post
I've made some progress getting better voices on my Ubuntu install.

I downloaded Amy from http://www.cepstral.com/en/personal and ran the installer, then by changing speak.sh in my homeseer folder to:

#!/bin/sh
#flite -voice kal16 -t "$1"
#pico2wave -w=temp.wav "$1"
#aplay temp.wav
padsp /opt/swift/bin/swift -n Amy "$1"

Amy now speaks but complains that she is not licensed. I'm in contact with Cepstral to see how to remove that. It looks like they have a Linux arm version as well.
Nice! Is that a cloud solution or offline?

Edit: Ah, had a quick look at the website and I see it is an offline solution. Looks really easy to install now that you've found the syntax for the speak.sh file. Nice that there is a version for 64 bit Linux which is quite enticing since this would be a lower resource solution that the SAPI route. I note that a voice for Windows is $35, so guess it would be similar for Linux. Hmmmm ... Let us know what they quote you for Linux please.

Last edited by abwyatt; December 20th, 2017 at 10:48 AM.
Reply With Quote
  #40  
Old December 20th, 2017, 10:32 AM
abwyatt's Avatar
abwyatt abwyatt is offline
Seer Deluxe
 
Join Date: May 2016
Location: Ho Chi Minh City, Vietnam
Posts: 206
Quote:
Originally Posted by Pete View Post
Yeah better to copy the files you see above in the JPG of the speaker directory.

Originally here started with the entire Homeseer directory then deleted files not needed.

IE: just install Homeseer 3 for Windows on your VB, then delete all of the extra files in the directory while concurrently keeping what you need for running HS3.

Also make sure you have installed the dot net stuff on your VB. Thinking Dot Net 3.5 should suffice.

Thinking if you install HS3 it will install all of the dot net stuff automagically.

Soon you will be letting your fingers do the walking relating to this stuff.....
OK, clued in now. Install HS3 on the VB. Think it could need .Net 4 after having a peek in the config file.

Getting a bit late here. I'll continue on the weekend. On the road again tomorrow and Friday. Thanks again for your patience in guiding me through.
Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Linux TTS - Voices petez69 HS3-Pi Discussion 10 July 2nd, 2017 05:57 PM
Current Linux build? New beta Z-Wave Linux 216 doesn't load AbMagFAb HS3/PRO Beta Releases 5 April 4th, 2015 11:40 PM
Static voices when switching voices. stevene HomeSeer Voices 0 July 22nd, 2007 08:16 PM
NeoSpeech and Asian voices / Fun voices GenevaDude HomeSeer Voices 5 December 29th, 2003 08:16 AM


All times are GMT -4. The time now is 09:25 AM.


Copyright HomeSeer Technologies, LLC