moz4r's blog

MRL cloud

moz4r's picture

MMMmm, hi ! Just an idea to discute about.

MRLcloudServer can be used , exemple to :

  • USE MULTIPLE TTS WEBSERVICES TO CACHE ( high availability and better lantency )
  • OFFICIAL TTS SERVER
  • SHARED KNOWLEDGE
  • VERSION CHECKING ( if mrlclient or script is outdated and broken )

What do you think about this architecture about tts :

 

 


AudioPlay mouthcontrol MaryTTS

moz4r's picture

This is a tip to have MRL mouthcontrol worky with external mp3 and maryTTS

( waiting something cleaner : other acapela service or maryTTS audioplay or mouthcontrol in audioplay )

add this at the top of your script :

AudioFile = Runtime.createAndStart("AudioFile", "AudioFile")

and this is to play the sound with mouthcontrol


MaryTTS - go further [post can be deleted]

moz4r's picture

[don't know how to delete the post]

--------------------

This post to list what we can do with the MaryTTS service in MRL.

I just discover it, there is a lot of interresting things

AudioEffets ( great tool to tweak the voice ) :

mouth.setAudioEffects() ->

Robot
Whisper
F0Scale
F0Add
Rate
Whisper
Stadium
Chorus
FIRFilter
JetPilot

Download voices

installSelectedLanguagesAndVoices() ->


Marty the robot powered by Myrobotlab !

moz4r's picture

Marty using MRL inside, great ! I'm using raspi + Adafruit16CServoDriver only.

 

 

 

Small setup :


openweathermap Weather API

moz4r's picture

updated this post to use openweathermap service, instead of python script

[ [service/OpenWeatherMap.py] ]

https://github.com/MyRobotLab/pyrobotlab/blob/develop/service/OpenWeathe...


Inmoov OpenCV test

moz4r's picture

Hi all ! I'm starting working on opencv, I need to read lot of documentations of course,it is so big.  But is the method I used is the way to do ?

 

I use the main loop of published onOpenCVData and intercept the boundarybox data of the facefilter


BOT memory

moz4r's picture

Hi this is what I use to give the chatbot some learning capacities:

It's a WIP

3 things:

Short time memory:

I used <topic> tag and <set> vars

They are temporary informations that are used only, example to turn on specific subject of conversation

<think><set name="topic">SPEAKABOUTROBOTS</set></think>

Set the conversation subject to SPEAKABOUTROBOTS

So all the AIML into


Inmoov messenger

moz4r's picture

Little update of the bot, this is inmoov ( and friends ) messenger .  Sorry for translation quality of messenger AIML. You can mod it .

 

 


WebkitSpeechRecognition works great !

moz4r's picture

Thanks the team . i can now stop the listening and resume it and never touch the microphone icon

Robot is now autonomous

I removed the microphone from the head to avoid servo noise ,  i put it later into coton away from servo.