BigTalker 2.0 -- Development

Got it … found it on code too … I should look at it first before asking but many thanks to you and others … I guess I have to buy a cheap speakers instead of LanNouncer

    > if (state.speechDeviceType == "capability.musicPlayer") {
        section ("Adjust volume during announcement (optional; Supports: Sonos, VLC-Thing):"){
        	input "speechMinimumVolume", "number", title: "Minimum volume for announcement (0-100%, Default: 50%):", required: false
            input "speechVolume", "number", title: "Set volume during announcement (0-100%):", required: false
            input "speechVoice", "enum", title: "Select voice:", options: state.supportedVoices, required: true, defaultValue: "Salli(en-us)"

Well the strange voice was do to latency, so that is solved.

NEW PROBLEM!!! When I have a room and say example “Turn off Family Lights” The spoken phrase is repeated twice, but the second is slightly different. Also I changed the phrase in the room that has two devices to a custom phrase “Family Room Lights are now off or on” Is it a bug in personality? When I use the toggle to test the phrase it repeats twice. Also in another room with two switches also when I tell Alexa “Turn on kitchen lights” the custom phrase is Kitchen lights are now on or off. But it says Kitchen lights are now off, then Dinette is now off. So something is screwed up.

I tested with mine (P2.0.b3, C2.0.b5) using a group that contained a switch event which contained multiple switches.
I could not duplicate the issue.
Are you sure you do not have your device in multiple groups causing triggering in each?

Please ensure that you have debug logging turned on and then pull up the logs in the API. Toggle the switch to test a phrase and PM me the logs from the parent and child app(has the name that you named the group and may not have any logs as I tend to send those to the parent apps logging).

In the PM, please highlight all of the pasted logs and click </> in the post editor to make them easier to read.

Is there a code for big talker to tell the time once when motion is detected when entering a room? For example when in away mode and it changes back to home mode enter living room big talker say the time is now %time% only once.

BigTalker can tell you the time as the phrase or part of the phrase when an event occurs, just insert %time% in the phrase.
It currently does not have a mechanism to restrict that to only x number of times per day though.

P2.0.b4 released to the Development branch.

  • Resolves an issue where Talk() is passed a blank phrase and processPhraseVariables() is called incorrectly causing an app crash. (Thanks for reporting that @greg)

Update from GitHub
or from the SmartThings IDE using GitHub Integration.

Publish the parent. The child app has not been modified it is still at C2.0.b5.


@rayzurbock Brian I am trying to eliminate problems where there are rooms with more than one device and there are multiple verbal expressions. I have tried almost everything including removing devices from a group. Is is possible that because I have groups in other places other than Big Talker, causing this?

Excuse my ignorance, but there is no BigTalker app in App Store. Is this Android only?

Grab it from his git hub.

And install with these directions

Thank you for the links. Onward!

BigTalker is Genius! Having lots of fun with this…

Most of my spoken text begins with Excuse Me Sir The Nest Thermostat is heating! The issue that i’m having is that by the time it plays thru my Apple MacMinis using VLC Thing, you only hear “The Nest Thermostat is heating” which isn’t terrible but it would be great to hear the entire message. Is there something I can adjust to make this work properly in order to hear the missing words in the beginning… otherwise this is still a stellar SmartApp! Thanks @rayzurbock

Simply insert Hey, or Hey Hey in front of Excuse…

I have inserted a word or two that I don’t care are skipped over or occasionally included as long as the bulk of what I wanted to make it thru the speakers is heard.

I have done this successfully with Big Talker, ES, AA.

I have the opposite problem, mine is cut off at the end. I think this is cloud delay processing and not Big talker.

I have never had an issue with audio being cut off at the end that is being sent to Bose speakers through any app or with any device. The delay up front is a known issue across the board with any external speaker (exception LANNouncer, etc), and that’s what the majority of people using voice are experiencing from what I see posted throughout the community. You have speakers (other thread) that are automatically decreasing in volume so I would say what is happening in your environment is unique and the minority. Are you sure it’s cutoff or that by the time it gets to the end of the message the volume is too low? :slight_smile:

In theory you could add text to the end and try it the same way it works at the beginning and see if it works for you.

1 Like

I am sure it is the DTH as I can see it happening. And no the clipping isn’t from the volume, as I hear a word cut in two. My problem is with long statements, not three four or even five words, but around six.

1 Like

Well, keep things short and tight and avoid the issue. :slight_smile:

1 Like

I see what you mean. I too am experiencing the cut off at the end with certain phrases.

1 Like

That is exactly what I have done, and am testing length of statements. By the way this happens also with WebCoRE pistons using speech as well as other apps. So keeping it short goes a long way to make this work perfectly.

1 Like

Please let me know what this do “Allow personality”?

It adds some uh, well, humor to the text being spoken. :slight_smile:

I didn’t mess with it much so I don’t know how much variety the dev has added but it adds things similar to ‘All I Know Is’… and then the rest of the expected text to read.

It’s a fun feature although I don’t expect I’d use it beyond testing. I kept getting odd looks from my wife since she thought I was adding the ‘personality’ to it.