Hi there,

I have a client which is creating e-learning, primarily for smartphones in regions of the world where internet is scares, and often unavailable.

So, I was wondering - if it would be possible to create interactive experiences with Construct 3, that can run in the browser also when the device is offline?

Hoping to hear back from you all :-D

Best wishes

/Thomas

Hi there,

thank you so much for pointing me in the right direction!

It was, as you said, the function not returning a value which caused the undefined value error.

I just added "return clean;" to the script and now everything works as expected :-D

Thank you so much for helping out!

Hi there,

I have two global variables and a script.

What I am trying to do is the following:

1) Pass the first global to the script

var result = showMessage(runtime.globalVars.speechFinal);

2) Set the second global with the result

runtime.globalVars.finalOutput = result;

The script has the following and delivers the expected output in the console:

function showMessage(interimresult)

{

var string = interimresult;

var cleanSet = new Set(string.split(' '));

var clean = Array.from(cleanSet).join(' ')

console.log(clean);

clean.toString();

}

However the second step is failing.

When trying to do this I get the following error:

Cannot read properties of undefined (reading 'toString')

Any help or pointers would be greatly appriciated!

Best wishes

/Thomas

Hi there,

nope I get the "Speech synthesis is supported" message.

But when I press the Speak! button, nothing happens

Hey dop2000,

thank you so much for helping out!

I just tried the official Speech Synthesis template on iOS - and get the same result....which is silenvce ;-)

For reference I uploaded it here: didaktik.digital/test

Do you think it is a bug we should report?

Best wishes

/Thomas

Hi there,

for some strange reason the webspeech api works perfectly in safari on my iOS device with this codepen example: codepen.io/matt-west/pen/DpmMgE

However, when I use the exact same settings with the Construct tts plugin nothing happens. Just get silence.....

I am not sure what I am doing wrong?

Any help would be greatly appriciated.

Best wishes

/Thomas

Hi there in this wonderful community :-D

We have a game where we are using the SpeechRecognition Plugin, which works as intended.

Except on iOS (which is not a Constrcut limitation, but an iOS thing.)

To fix this we would like to try out this solution from our forum member Hodgekins (https://www.construct.net/en/forum/construct-3/how-do-i-8/speech-recognition-ios-devices-148527)

As stated this is a paid oppurtunity - and we would love to find someone we could keep continue working with.

Hoping to hear from you!

Best wishes

/Thomas, Didaktik.digital

Kiel, Germany

Hi there,

I have a small project where I use the SpeechRecognition Plugin and on the start of the project ( on the login screen) I check if SpeechRecognition is supported.

Now, on Chromes as well as Edge and Firefox this returns "Mic is working".

So what I then do ingame is, I check the SpeechRecognitionResult against a string to see if they match.

However, the outcome are compleatly different, depending on the browser:

Chrome: If SpeechRecogntion == Result -> Yay, they match.

Edge: If SpeechRecogntion == Result -> I am sorry they don't match.

Firefox: Never returns any result at all.

So, it looks like Edge is reversing the result and Firefox always returns a blank.

My question is:

Am I doing something worng in the intial check?

I would really like to have a valid check wetaher or not the browser is capable of using SpeechRecognition, as intended.

Best Wishes

/Thomas

Hi there,

if I understand the STT Plugin correctly it uses the webspeech api, which in turn uses google to process the respons. This approach may cause some concern in our use case (education).

So I was wondering if it would be possible to send the STT input from the user to our own server, and then from there pass it on to google for processing. Basicly proxying the input so the user stays anonymous.

I was thining about sending the audio via WebSocket to a construct app on our server, which then does the STT and returns the string. But to be honest I have no idear if this is even the right place to start.

Any help would be greatly appriciated, even just theoretical thoughts :-D

Thank you and best wishes to all of you!

/Thomas