Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Brandt]

Hey all,

As some of you may know my house is powered by Linux 

Anyway, I've been trying out the Linux text-to-speech options and they are far behind windows and mac, they sounds like robots...Also, the system I'm running on (1.2GHz, 512MB) has a hard time doing TTS in a timely manner, there is an annoying delay before it starts talking (like a robot)

So I've heard a bunch of really nice TTS voices of which none are available to me. Therefore I've decided to go with pre-recorded TTS voices generated from my Mac, and a hybrid of native TTS with pre-recorded as well.

Then I copy the sound files over to my linux machine into a sounds folder. I break them down into re-usable phrases like; "excuse me", "thank you", "goodnight", "good morning", and a recording of everybody's name in the house.

If you're interested in the command line syntax on a mac it is:

Code: [Select]

say "excuse me," -o excuseme.aiff
Then once transferred to my linux machine, I have them play in macros/triggers/scripts whatever you want to call them like this:

Code: [Select]

play -q excuseme.aiff brandt.aiff dinner.aiff thankyou.aiff
The "-q" is so the play command doesn't spit out any unnecessary text on the command line since I run it headless, don't need it to show up in the logs.

Now if there is some text that needs to be generated on the fly like the temperature (not going to pre-record hundreds of numbers) I have to do it like this:

Code: [Select]

flite -voice slt $TEMPERATURE tmp.aiff &; play -q excuseme.aiff brandt.aiff temperature.aiff tmp.aiff thankyou.aiff; rm tmp.aiff
This uses flite TTS program to generate a temp sound file called tmp.aiff from the variable $TEMPERATURE with the voice "slt" and run in the background (&). Then play will start right away with the pre-recorded voices, will insert the tmp.aiff sound file which sounds a bit different but both are female voices. Lastly it will delete the temp sound file.


Now, my question to you is; what pre-recorded voice sound phrases are you using? I'm trying to build up my library....


 

[Tuicemen]

I had many wav files created for AlertDialer before I got it coded it to do TTS.
Some were for:

• Security Alerts
• Mail
• greetings (which you have)
• Different ringers for phone rings (modem needed)
• Secial events or times
• door bell sound ( tripped by a sensor at the front door)

[Brandt]

Thanks, I still need to do security alerts. I'm not monitoring mail or telephone lines. I may do the calendar reminders. Right now it is mainly being used with my RCS TXB16 to alert of open doors and windows, which will also be extended to the security system.

[HA Dave]

Now, my question to you is; what pre-recorded voice sound phrases are you using? I'm trying to build up my library....

My favorite is the one generated when one of my Floodlights (PR511) detects motion: "Activity has been detected! This is a restricted area. All activity in this area is off-site video monitored".
Also all my cameras are announced when turned on: "Nightvision camera is On".
My garage door reminder: "The garage door is open".
When my DS7000 is triggered and the console sends an All lights On (then all units Off) the All lights On causes the BVC setup to say: "An intruder has been detected".

Much of theis stuff can be seen on my YouTube videos

[Tuicemen]

I guess since you can and are making sound files you can do it for any thing PcCompanion or X10WavPlayer can say!
I use PcCompanion to annouce when most modules come on like Dave.
As well you could create a sound file for each hour and have a talking clock.
 

[HA Dave]

............ As well you could create a sound file for each hour and have a talking clock.

I forgot to mention my talking clock. It only uses twelve [X10 module] addresses as I don't use AM or PM.