Voice Recognition extension without Google Dialogue


Is a pure activation via audio command possible?
I have programmed an AI which can send questions via text and returns the text and also addresses the ESP8266 (poor and face) only I have to press the belly of the robot to activate the whole thing.
A keyword (like “alexa” from Amazon) that wakes the robot would be my dream so everyone could give his robot a different name.

Thank you for taking your time.


Using my extension, you can easily implement this.
The robot launches Continuously Recognition and waits for a keyword. When he hears him, he can answer “Yes, I listen.” He will take the next phrase as a command. If the command has not been heard for a long time, the robot will again go into the standby of the key phrase.
Implement this algorithm. I will help you if is nesecary.
You should also bear in mind that the Continuous Recognition can drain the smartphone battery in a short time. Because there is a constant audio recording. And, with online recognition, there will be a constant big traffic. You can give preference to offline mode, to wait for a key phrase, and switch to online, to recognize the command.


Thankyou very much!! Very useful for me.


very amazing, I can use it for learning in my school. thanks.


But I think we can use a good looking animation bar which is indicated when we’re recording the audio.

Exactly how google assistant does !!!

Is that to not allowed ?

-/ Aravind Chowdary
Know how to make your own extensions : click here


OMG PEOPLE YOURE NOT UNDERSTAND!!! You can record someone’s actions without needing them to know you are. You don’t need a button to start recording, ITS A BLOCK YOU COULD PUT UNDER SCREEN.INIT! You don’t even have to alert them that it’s happening.


This is shouting and we don’t shout here, so don’t do that. I already told you we stopped the discussion about, if you use this for good or bad so don’t start it again.


Standard component SaundRecorder too. I agree with @Peter. The discussion is closed.
Merry Christmas!


Updated extension.

New in the Version 3:

New Properties KeepScreenOn (Boolean, default is False) - If you app window is visible to the user, keep the device’s screen turned on and bright

New Event OnRmsChanged Event occurs when Microphone input is change. Gives an integer 0…10. It is used to dynamically indicate the microphone status in the voice recognition process.

You can download in the first post.


I would love to come back to your help.
I bring this with the offline listening and after the keyword online just don’t go there.
Do you have an example code?

Thank you very much for your help.


Sorry. I do not quite understand what is the problem.


Olá, sou do Brasil. Você poderia me explicar melhor o que a sua extensão faz? Eu não entendi muito bem. Tem como eu deixar o App escutando meus comandos de voz a todo tempo, sem eu precisar apertar nenhum botão? Se puder me responder eu agradeço. Valeu!


Hola, Leo.
Si, tu puedes organisar dialogo sin nesecidad de tocar la pantalla.
En inicio de aplicacion:
Tienes que ajustar ContinuouslyRecognitionEnabled=true ,
Despues GetText() puedes decir algo (extension va esperar hasta cuando tu dices algo),
El evento AfterGetting(Text result) va volver te un resultado de reconocimiento,
Puedes hacer algo con el resultado y volver pedir GetText()



if i use that extension and i want to upload my app in play store, do i get a probem with google?


The extension is made on the basis of open Android API. There should be no problems. But it is better to read all agreements with Google.


Olá, mesmo eu desativando o som do dialogo do google, ele continua fazendo o bip, pode me ajudar?


Hola, el bip puedes bloquear solo en caso si no estas usando dialogo Google:


Salutations Mr. @ScorpionNormal , Thank you very much for this extension…
Please do you have an example code?


The example from the first post demonstrates all the features. If you need more, I will try to help.


Great work, very useful.
at the end of the aftergettingtext block I tried to add the
call-scSpeechrecognizer1.getText to get a loop, a continuous listening even after the first text recognition. The problem is that, although the MuteCheckBox is checked, beeps are heard at the beginning and at the end of the recording.
Any suggestions? Thank you very much.